The Venture Of Islam Clasical Age Vol 1 Marshall Hodgson

The \enture of Islam

Conscience and History in a World Civilization

The Classical Age of Islam

Marshall G. S. Hodgson

■y>.j

\0'

►

T*

-

rJ 1

.'. r>

SET^

*■*

1U

J

The Venture of Islam

To consider mankind otherwise than brethren,
to think favours are peculiar to one nation
and exclude others, plainly supposes a
darkness in the understanding.

— John Woolman

The Venture
of Islam

Conscience and History in a
World Civilization

MARSHALL G. S. HODGSON

VOLUME ONE
THE CLASSICAL AGE OF ISLAM

THE UNIVERSITY OF CHICAGO PRESS

CHICAGO AND LONDON

To John U. Nef

and to the memory of

Gustave E. von Grunebaum

in admiration and gratitude

Some of the material in these volumes has been
issued in a different form in Introduction to
Islamic Civilization (volumes i, 2, 3, Copyright ©
1958, 1959 by The University of Chicago), in A
History of Islamic Civilization (Copyright © 1958
by Marshall G. S. Hodgson), and in an earlier
version of The Venture of Islam (volumes 1, 2,
Copyright © 1961 by Marshall G. S. Hodgson).

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 1974 by The University of Chicago

All rights reserved. Published 1974

Paperback edition 1977

Printed in the United States of America

ISBN: Vol. 1, 0-226-34683-8 (paper);

Vol. 2, 0-226-34684-6 (paper);

Vol. 3, 0-226-34681-1 (cloth); 0-226-34685-4 (paper)

Library of Congress Catalog Card Number: 73-87243

03 02 01 00 99 98 97 10 11 12 13

© The paper used in this publication meets the
minimum requirements of the American National
Standard for Information Sciences — Permanence of
Paper for Printed Library Materials, ANSI Z39.48-1984.

CONTENTS

VOLUME I

List of Charts vi

List of Maps vii
Marshall Hodgson and The Venture of Islam, by Reuben

W. Smith viii

Publisher's Note xi

Introduction to the Study of Islamic Civilization 3

General Prologue : The Islamic Vision in Religion and in Civilization 71

Book One: The Islamic Infusion: Genesis of a New Social Order 101

I. The World before Islam 103

II. Muhammad's Challenge, 570-624 146

III. The Early Muslim State, 625-692 187

Book Two: The Classical Civilization of the High Caliphate 231

Prologue to Book Two 233

I. The Islamic Opposition, 692-750 241

II. The Absolutism in Flower, 750-813 280

III. The Shar'i Islamic Vision, c. 750-945 315

IV. Muslim Personal Piety: Confrontations with History and

with Selfhood, c. 750-945 359

V. Speculation: Falsafah and Kalam, c. 750-945 410

VI. Adab: The Bloom of Arabic Literary Culture, c. 813-945 444

VII. The Dissipation of the Absolutist Tradition, 813-945 473

A Selective Bibliography for Further Reading 497

Glossary of Selected Terms and Names 513

Index to Volume I 519

CHARTS

VOLUME I

Book One

Twentieth-Century World Distribution of Muslims 76

Overview of the History of Islamdom 96

The Place of Islam in the Chronology of World History 113
The Development of Confessional Religion in the Irano-

Mediterranean Region, c. 650 BC-632 ce 126-27

Cultural and Religious Orientations between Nile and Oxus 131

The Origins of Islamic Culture in Its World Context, 226-715 139

Muhammad's Relatives 168
Chronology of Muhammad's Lifetime 188-89

Chronology: Abil-Bakr to 'Abd-al-Malik, 632-692 200

The Public Cult as It Had Developed by Marwani Times 210

Events of the First Fitnah, 656-661 215

Events of the Second Fitnah, 680-692 220
The Umayyad Caliphs 224-25

The Tribal Blocs of the Garrison Towns, by Alleged Genealogy 228

Book Two

Comparative Periodization of the Caliphate to 1258 234

Chronology of the Marwani Umayyads, 692-750 244
The Main Khariji (Shurat) Movements through the Time of the

Third Fitnah 257

The Candidates of the Primitive Shi'ah 261

Piety-Minded Groups in Marwani Times 263

Events of the Third Fitnah, 744-750 277

The Early 'Abbasid Caliphs 285

The Flowering of the High Caliphate, 750-813 288

The Masters of Fiqh 319

The Derivation of a Shar'i Legal Decision 338

The Ja'fari and Zaydi Shi'ahs 375

The Earlier Sufi Masters 408

Early Islamicate Philosophic Schools 434

Earlier Translators, Philosophers, Scientists 437

vi

CHARTS Vll

The Earlier Classical Arabic Belles-Lettrists 446
The Weakening of the High Caliphate, 813-945 476-77

The Fourth Fitnah: al-Ma'mun's Wars 479

The 'Abbasid Caliphs, 833-945 484
Successor States and Principalities: The Weakening of Caliphal

Control 489

MAPS

VOLUME I

Countries of the Afro-Eurasian landmass, 1970 84

The central Mediterranean through India, c. 600 CE 116

Trade routes of the seventh and eighth centuries ce 122

The lands from Nile to Oxus before the rise of Islam 129

Towns and tribes in Arabia in the time of Muhammad 152

Conquest of the Fertile Crescent and adjacent lands to 656 202

Muslim lands at the time of Harun al-Rashid, 786-809 293

Muslim lands, ninth and tenth centuries 311

Marshall Hodgson and The Venture of Islam

Marshall Goodwin Simms Hodgson died suddenly on June 10, 1968, in his
forty-seventh year, before he had finished this and other works. At the time
of his death the manuscript for the first four of the six books of The Venture
of Islam had been submitted to a publisher as final (although minor editing
and footnote attribution still remained to be done); he had reworked much
of Book Five; and he had indicated revisions wanted in Book Six. Many of
the charts and diagrams were only sketches, and there were only general
descriptions of the intended maps. A much shorter version of the work,
resembling The Venture of Islam in its general form, had existed for a
number of years, used and sought after by graduate and undergraduate
students alike. It had started as a brief outline, some chapters consisting of
only two or three pages or not yet even written. He constantly wrote,
withdrew what he had written, rewrote, and sought criticism from his
colleagues near and far. He was simultaneously at work on a world history,
and he often remarked that he hoped readers would some day have both
available, for he had long been convinced that any historical development
could only be understood in terms of the historical whole, and that whole
included the entire canvas of human history. Although several hundred
pages of manuscript for his world history exist, most unfortunately it
apparently cannot be published, for to put it in meaningful form would mean
that it was no longer his work. A number of the world history's basic
assumptions and points of view can be found in The Venture of Islam, how-
ever, particularly in the various sections of the 'Introduction to the Study
of Islamic Civilization', originally planned by Hodgson as appendices to
the entire work. He was an indefatigable and fastidious worker, with
definite ideas of his own. Although continually seeking advice from every-
where, he was yet adamant that no publisher's editor would alter his
text.

It was with considerable trepidation that I agreed to see this work through
the press after his death. I had worked with him, even sharing his office
for a time, and had taken over the course in the history of Islamic civiliza-
tion at the University of Chicago, a course that he had created and that we
had for two years jointly taught. The Venture of Islam had originated to
meet the needs of that course; but concurrently Hodgson recognized that
much he had to say went far beyond any ordinary text. He always hoped
the book would appeal to the educated layman as well as to the specialist
and the beginning student; he thought he could reach a'' three in the one
work, and it was thus that he wrote.

viii

MARSHALL HODGSON IX

He kept voluminous notes, and he had written out for himself many
directions; there were numerous complete charts as well as sketches and
designs for others. The same was true for illustrations, which unfortunately
because of cost have had to be eliminated. I saw my task to be as light-
handed as possible, and to preserve the manuscript as completely his.
Therefore, there is a more detailed text, and there are more charts and maps
for the first two-thirds of the work than for the last third. His writing style,
particularly, is unaltered; there were places in the manuscript where he had
deliberately changed to a greater complexity from an earlier, less detailed
style. Always his aim was to pack as much meaning as possible into a
sentence or a paragraph, while keeping it so circumscribed as to include only
what he wanted. Some of his neologisms may not be attractive; more than
most other writers, however, he has made his readers aware of the traps
one may fall into when giving a word or concept familiar in one culture an
apparently similar connotation in the context of another culture. His
world history would have had central to it such concerns. I followed his notes
where I could. Since he opted for different spellings and even in some cases
different terminology later in his writing, I have tried to bring some con-
sistency there. In a few instances he altered traditional dates to other, less
usual chronology; these dates I have left. Only where a fpw notes in
brackets are found in Book Six has my presence intruded. I hope and
believe the work is his, purely and directly, and that I have done nothing
to alter it.

No one who was associated with Marshall Hodgson remained unmoved by
who he was and the scholarship he stood for. He was a lesser-known giant
among better-known scholars. His Quaker background provided him with a
quiet gentleness backed by absolute resolution when necessary; perhaps
more than for most teachers, the kind of person he was informed the classes
he taught, especially those in Qur'an and Sufism. No narrow specialist, he
found in his work with the intellectually wide-ranging Committee on Social
Thought at the University of Chicago a satisfaction rivaling that afforded
by his Islamic study and teaching.

It is impossible here to make the acknowledgments he would have made
if he were alive. Leading Islamic scholars throughout the world read portions
of his work. He was able to take few leaves from his teaching and his
administrative duties to devote entirely to research; but one or two precious
opportunities did occur. Help also came from friends, and especially his
students. I would like to acknowledge simply by name help graciously and
eagerly offered me by his colleagues and students: Professors William H.
McNeill, Muhsin Mahdi, and the late Gustave von Grunebaum for valuable
counsel; graduate students Harold Rogers, Marilyn Robinson Waldman,
William Ochserwald, and George Chadwick, now all launched on careers
of their own; and many others too numerous to mention.

Most of all, the countless hours Marshall Hodgson's wife, Phyllis, devoted

X MARSHALL HODGSON

to the whole work over the years under the most trying of circumstances
cannot be left unacknowledged. Her example, as his, remains for the rest of
us a monument.

Reuben W. Smith
Callison College
University of the Pacific

Publisher's Note

Reuben Smith's account of his own contribution to the posthumous
realization of Marshall Hodgson's monumental work is far too modest.
Over a period of at least four years he gave uncounted hours, which he might
otherwise have spent on his own research and writing, to the difficult task
of finishing his colleague's work. For this selfless devotion, and for his never-
failing good humor in answering countless questions, the University of
Chicago Press wishes to thank Mr. Smith on behalf of all students and other
readers of these volumes.

When the University of Chicago Press acquired The Venture of Islam
from a commercial publisher, all the text had been set in type by Unwin
Brothers Limited in England and galleys had been proofread by Mr. Smith.
Decisions remained to be made concerning the charts and the maps. After
consultation with Mr. Smith, the Press decided to include all the charts
originally planned by Marshall Hodgson except a very few incomplete or
peripheral ones. The titles of the charts are Hodgson's, as are the concepts
and the contents, as nearly as we could reproduce them.

A tentative list of map titles had been drawn up by Smith from Hodgson's
notes and references in the text. Using this list and drawing on his wide
knowledge of the field as well as his understanding of Hodgson's aims and
point of view, John E. Woods, of the Department of History of the Univer-
sity of Chicago and currently one of the professors teaching the course in
Islamic civilization, drafted all the maps included in these volumes. From
his drafts, the Cartographic Division of the University of Wisconsin provided
the final versions.

The indexes to all three volumes were compiled by Yusuf I. Ghaznavi,
graduate student at the University of Chicago, with the assistance of his
wife, Huricihan, a former student of Hodgson. Mr. Ghaznavi also adapted
Hodgson's original Glossary to the three-volume format so that the user of
each volume would have definitions of the unfamiliar words at hand.

The decision to make these volumes available singly as well as in a set
necessitated a few changes from Hodgson's original plan. In a series of
'Appendices' Hodgson had set forth his views on the meaning of Islamic
terms, transliteration, personal names, dates, historical method, and other
important aspects of his work. Since familiarity with this material enhances
the reader's understanding of the entire work, we decided to move it all to
the beginning of Volume I, changing only the title, from 'Appendices' to
'Introduction to the Study of Islamic Civilization'. Cross references in the
text to the 'Appendices' have (we hope all of them) been changed to refer
to the Introduction. Hodgson conveniently arranged his 'Selective Biblio-

xi

Xll PUBLISHER S NOTE

graphy for Further Reading' in chronological sections that enabled us to
divide it into the three relevant parts, and accordingly, each volume has its
own 'Selective Bibliography'. General works are all listed in Volume I.
The question of the Glossary was more difficult. Convenience of the reader
again indicated the desirability of a glossary in each volume, even though
many words would appear in all three. Finally, while Hodgson supplied
titles for each of his six 'books' and for the work as a whole, he did not
envisage its being divided into three volumes; from terminology he used
elsewhere, we have provided titles for the separate volumes.

The University of Chicago Press is pleased to be able to bring The Venture
of Islam to the many students and friends who have long awaited its
publication and regrets only that Marshall Hodgson himself is no longer here
to share that pleasure.

The Venture
of Islam

Introduction to the
Study of Islamic Civilization

On making sense of Islamicate words, names, and dates

The thoughts of a people distant in time or space cannot be at all deeply
shared without our becoming acquainted with things and ideas important
to them but of which we have had no exact equivalent. As far as possible,
one wants to read the works themselves in which the thoughts have been
expressed; in these, even in translation, the special concepts and categories
of the writers, as well as the personalities and places referred to, must be
reproduced (if the translation is serious) in forms alien to the usual flow of
English, no matter how much the resources of English may have been
adapted or even twisted to do duty for what remain alien conceptions.

The same is, in some degree, true of any work treating of the alien civiliza-
tion. The serious reader must be prepared to think in novel ways. To this
end, he must be prepared to absorb as readily as possible a whole range of
new concepts and terms. Otherwise he cannot expect to profit seriously by a
study of the culture; at most he will receive an impression of exotic quaint-
ness, romance, or incongruity which does no justice to the human reality.

Though Islamicate culture has been expressed in many languages essentially
unrelated to each other, much terminology and customary practice has
been common to them all. For instance, technical terms in religion and also
in some other fields have commonly been derived from the Arabic or in
some cases from the Persian, as have been Western terms from Latin and
Greek. Just as is the case with Christian names, Muslim names form to a
large degree a common stock that reappears substantially the same in every
Muslim country. The manners of dating an event or of heading a letter
tend to be constant, and of course the use of Arabic script. It is important
to feel as much at ease as possible with all this.

The problem is complicated by the fact that in many cases writers about
Islamdom as well as translators have been very inconsistent in their render-
ings of names and concepts. The reader will find the same term presented in
many utterly different guises. The various sections of this Introduction give
a number of aids for negotiating the resultant maze. Ways of transliterating
from various Islamicate languages are outlined, with suggestions of how the
reader can refer from one system to another if he reads different authors;
Muslim personal names are grouped into common types, with suggestions
for keeping them apart; the Muslim calendar is explained; short essays are
offered on problems of studying a civilization ; recurrent technical terms are
briefly defined. Leaving most detail to the following special sections, I
believe it necessary at this point to emphasize reasons for using exact

3

4 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

transliteration, and to offer some suggestions of how the systems used in this
work may at the same time assist approximate pronunciation.

Why transliteration?

Transliteration is the rendering of the spelling of a word from the script
of one language into that of another, in this case from the 'Arabic' alphabets
used in Islamicate languages into the 'Latin' alphabet used in English.
Transcription is the rendering of the sound of a word so that a reader can
pronounce it. Ideally, transliteration should include as much transcription
as possible. When an original script is unfamiliar, transliteration is necessary
first of all in the case of names, which by their nature cannot be translated
and yet must be clearly recognizable and distinguishable from any names
like them; and secondly in the case of technical terms, when the concept
they represent is not present in another language and it is important to
refer to it with precision.

In this second case of technical terms there is room for difference in the
degree to which transliteration is used as compared with a rough translation.
Different writers draw the line at different points. A word that some will
transliterate, others will try to translate by more or less equivalent English
words, either because they attach less importance to precision in that given
case, or because they hope to attach a technical meaning to a special English
word or pair of words. Thus the concept expressed in all Islamicate languages
as xjlj yZ, is not present in English. If it is translated as 'law' the
reader is misled, because it covers much that we do not call 'law' and fails
to cover much that we do call 'law'. Even if we use a compound term (as is
often advisable), 'Sacred Law', the reader will be misled unless he is given
a full explanation that this is something quite different from what the
term 'sacred law' might ordinarily lead him to expect. In The Venture of
Islam the concept here being referred to has a special importance; hence
for exactness it is rendered, according to its sound in Arabic, by the trans-
literation (which is also a close transcription), Shari'ah, and it is explained
in detail. The like is done with a number of other important terms that
occur frequently. In this way, so far as a term carries any meaning at all,
it will be that assigned it in the explanation and not one derived from over-
tones attaching to previously known English terms. In addition, it is un-
mistakable which original term is referred to; whereas arbitrary translations,
varying according to the writer, often leave the careful reader in doubt.

If a system of transliteration used for names and for terms is consistent
and exact it saves everyone much time and effort in the long run. Above
all, any system must be reversible; that is, such that the original written
symbols (which are far more constant than the pronunciation) can be
reconstructed with certainty from the transliteration. Second, it should so
far as possible at the same time be a transcription, such that the careful

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 5

reader will pronounce the word more or less recognizably in oral communica-
tion. Finally, it should require as few diacritics as possible so that, if they
are omitted, confusion will be held to a minimum; the diacritics that are
used should be common ones, available on the average academic or pro-
fessional copyist's typewriter.

Diacritics are, however, necessary. Since letters of one language usually
have no exact equivalent in the alphabet of another language, the choice
of English letters to render Arabic or Persian letters must be rather arbitrary.
Some alien letters can be rendered more or less happily by existing English
letters: for instance the Arabic «y sounds not unlike the English T. The
sound of others is unlike any sound in English; but they are usually in the
same phonetic category with some other sound, just as the G in gave is
related to the C in cave. For such we can sometimes use digraphs, like
TH for the Greek ©; but sometimes we must invent a new English letter
that is rather like an existing English letter of related sound. Thus for the
Arabic So we invent arbitrarily a new letter T, because to the native speaker
of English that Arabic letter is rather like T in sound. It must be noted that
T and T are quite as distinct from each other as C and G (in fact, G was
originally formed in the same way, simply by adding a stroke to C). The
added part — a dot, a stroke, or whatever — of the new letter is the diacritical
mark, and is sometimes omitted in printing (just as in certain Latin in-
scriptions both G and C are written as C, without the stroke). But properly
speaking, the diacritical mark is not something extra, to be omitted at
will ; it is an essential part of a new letter.

It might seem as if exact reconstruction of the original were of importance
only to the scholar who knows the original language. 1 In fact, it is of import-
ance to the ordinary reader also. He needs to know whether the scharia he
comes across in one writer is the same concept as the shareeah he finds in
another writer (it is) ; or whether Hassan and Hasan are likely to be one man
or two. {Hassan and Hasan are two quite different names; but newsmen,
who have no exact system, often write them both 'Hassan'.) That otherwise
educated persons seem to be helpless in the face of exact transliteration and
fail to profit by it does not show that on this point human nature is perverse
but only that Western education is remiss. The only way that this can begin
to be remedied is for scholars to set a good example and form better habits
in their readers.

1 It is sometimes said that specialists do not need exact transliteration and non-
specialists cannot use it. This is not so. Outside of a very narrow subspecialty in which
he has full first-hand acquaintance, even the Arabist or Persianist will come across
names first in secondary works, commonly in a Western language ; yet he will want to
know the original. Without diacritics he cannot distinguish 'All from 'Alt, Hakim from
Hakim, zahir from zdhir, 'Amir from Amir. To assume that the non-specialist can have
no use for exact transliteration suggests condescension; for it implies that he will read
only the one book on the subject, and so have no occasion to refer across from the usage
of one book to that of another, nor need to straighten out the names he comes across.

6 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

If all writers used a single consistent system, the problem of identifying
transliterated names would be much reduced. If that is impossible, then if
each author's system is at least exact, and he explains it (as careful authors
do) in a note or in his preface, a reference to the prefaces of the two works
will usually tell the reader whether or not one spelling is the equivalent of
the other.

Strictly speaking, it might be enough if the writer put the exact forms
only in the index, to avoid the cost of diacritics in the text. But this sacrifices
the advantage of accustoming the eye to the correct form rather than an
ambiguous one, as well as being a nuisance to that reader who wants to
read a passage uninterruptedly and yet be sure of references as he goes
along. Since careful writers now spell 'Cezanne', Tubingen', 'Saint-Saens',
'Charlotte Bronte', 'facade', 'Provencal', 'Potemkin', Arabic and Persian
words should also be spelled with diacritics. The careless reader should not
begrudge the careful one such precision. Nevertheless, in those cases where a
well-known name has already achieved a uniform rendering in English, it is
generally regarded as unnecessary to include it in the system of transliteration
since common Anglicized forms such as Cairo or Damascus — like Naples or
Quebec — are unambiguous. This is especially appropriate where the most
frequent context of the term in English will be other than special discussions
of Islamicate materials: in other words, situations that do not require
special new thought habits. A rough rule is that common English usage
should be retained when a term refers to something easily recognized and
presently existing, such as a great city. 2 (To allow for Anglicized accentuation,
diacritics may sometimes be dropped when an English termination is added
to words so common that they have already shown the exact spelling in the
simple form.)

2 It is worth noting that non-technical, 'popular' forms are subject to a steady
erosion, however much uniformity they may have achieved. Specialists will tend to
use the technical transliteration, first in special monographs and then by habit in their
general works. Once a majority of the specialists are using the technical transliteration
in general works, the rest tend to follow; and one generation later, the non-specialists
are likely to follow the usage of the specialists. Thus the name of the Prophet was once,
for non-specialists, usually Mahomet; this was almost universally replaced among
non-specialists by what used to be the technical transliteration of the specialists,
Mohammed; but meanwhile, a more sophisticated technical transliteration had
supervened among the specialists, Muhammad; now this (shorn of its diacritic) is
beginning to replace the older form also among non-specialists.

For those who would like to maximize continuity in the English literary tradition,
it is hard to know where to try to stem the tide. Once the technical form has become
the more common among specialists — e.g., Qur'dn for Koran — it is probably too late.
Very few terms regarding the past seem destined to retain a non-technical form, perhaps
only those prominent in European history and where the technical form requires a
major shift in pronunciation and not just in spelling [Ottoman for Osmanll, Caliph for
Khalifah). Current geographical terms, on the other hand, have a better chance despite
the tendency of some cartographers to put all names into native form, even Roma and
Napoli.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 7

Some general suggestions for pronunciation

Transliteration is required in this work chiefly from four languages, Arabic,
Persian, Ottoman Turkish, and Urdu, each of which has used a form of
what was originally the Arabic alphabet. Of these, Arabic words will most
often be met because, for the sake of uniformity, technical terms (whatever
the linguistic context) will be rendered only in a single form, that of their
original language, which is most often Arabic.

The letters used in this book for transliterating Arabic and Persian can
be pronounced recognizably enough by a beginner if he renders the vowels
as in Italian (a = ah, e as in bet, i = ee, o = oh, u = oo), and the con-
sonants in the commonest English way. Long vowels (a., i, u) should be
held extra long, and may receive the stress. The consonants must be pro-
nounced unvaryingly; thus s is always hissed, and h is always pronounced,
even in words like Mahdi or Allah (but not in digraphs like th, sh, ch
pronounced as in English). Beginners may pronounce the dotted or lined
letters like the undotted ones, but must keep them separate in the mind or
similar names may be confused. Aw is like ou in house; dh is th in then;
'{hamzah) and '('ayn) are consonants which are not always distinguishable to
the Anglophone ear and can be overlooked in pronouncing.

The same rules of thumb may also be used for the other languages, except
for Urdu (and Indo-Persian) , where a short a is like u in but; and a com-
bination of h with a consonant often is not a digraph but must be pronounced
as an aspirate; that is, th is not pronounced as in nothing, but somewhat
as in at home; and dh somewhat as in ad hominem. Ottoman Turkish names
may be treated like the Persian except that o and u are pronounced as in
German (French eu and u). Modern republican Turkish has a Latin alphabet
of its own, several letters of which must be learned ad hoc (see the section
following) ; but apart from these, the usual system will serve — vowels as in
Italian, consonants as in English. The same guide will serve at least roughly
for any other foreign names to be encountered.

It is wise for the beginner to use no stress at all, but to pronounce each
syllable with equal fullness (unless he knows that a given syllable is to be
stressed).

Certain Muslim names are often spelled in English in two portions, which
are, however, essentially inseparable. Beginners (and journalists) often drop
the first portion, treating the latter portion as a 'last name' to be used alone.
This is like referring to MacArthur as 'Arthur'. Abu-Yazid, for instance, is
a quite different figure from Yazid, Ibn-Sa'd from Sa'd, and 'Abd-al-Malik
from Malik. (Still worse than calling 'Abd-al-Malik 'Malik' is the comple-
mentary error of treating the ' 'Abd-al-' as if it were an independent 'first
name'; forms like 'Abdul', 'Abul', or TbnuF when thought of as separable
names are barbarisms that are not even formed in the original script, let
alone actually used.) In this work such compound names will be hyphenated ;

8 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

but many writers separate them without even a hyphen, and the beginner
may well learn to restore the hyphen mentally, to avoid endless confusion.

TRANSLITERATION

Transliteration from Islamicate languages, especially from Arabic, has been
increasingly uniform in scholarly works in English. Yet even now writers
differ; books written in the last century and the beginning of this century
show a great variety of systems (and lack of system). A serious reader who
goes beyond one or two books must gradually become used to the chief
older variations as well as the the main modern systems. 3

As suggested in the preceding section of this Introduction, there are
three practical requirements in transliteration: written reversibility, that is,
one should be able to reconstruct the original written form from the trans-
literated form; oral recognizability , that is, the reader's pronunciation should
be well enough guided to allow oral recognition by someone who pronounces
properly; and resistance to accident, that is, if diacritics are changed or lost,
either by chance or for economy, confusion should remain minimal. To
fill all three requirements, there must be a separate system from each language
into each language. Arabic and Persian, for instance, not only pronounce the
same letters very differently, but they are two different languages. Hence
they require not just distinguishing transcriptions but separate translitera-
tions. 4 Some scholars interested only in written reversibility have tried to
use a single system for transliteration into all languages using Latin script.

3 Detailed differences in transliteration continue to be multiplied. For instance,
Cowan, in editing the translation of Hans Wehr's Dictionary of Modern Written A rabic,
as late as 1961 introduced two new symbols. Variations are often not accompanied by
a justification, and they may only increase the general problem. By retaining, for
extraneous reasons, the old dj and k, the second edition of the Encyclopaedia of Islam
lost an opportunity to provide a final settlement of transliteration of Arabic, at least,
into English.

1 Once there was an attempt to transliterate all Semitic alphabets by one system
(Wellhausen) ; this included Arabic and by extension Persian, Turkish, etc. Then an
attempt was made to transliterate uniformly all languages using the Arabic alphabet.
In addition to the difficulty of maintaining some semblance of transcription for the
other Islamicate languages, such as Persian, 'uniform transliteration' foundered on the
fact that Urdu (Hindustani) was written in two alphabets (which were thus handled
differently) ; and either the Islamicist's way of writing Urdu had to look too clumsy in
comparison with the alternative, or his way of writing Arabic had to be so denatured
as not to stand comparison with the transliterations for other Semitic languages. But
there is an even more serious dimension when an alphabet-to-alphabet system instead
of a language-to-language system is pressed. If Persian or Turkish is felt as a language
in its own right, a 'uniform transliteration' point of view can become offensive, because
not only pronunciation but certain syntactical practices of Persian, for example, are
made to seem merely an 'exception' to the rules for Arabic. In a general work, such a
misapprehension must be avoided. Uniform transliteration implies more concern with
philology than with lay intelligibility. The philologian can easily learn to convert from
the system of one language to that of another, while the layman needs assistance barely
to pronounce.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 9

The problem is, the same Latin-script letter can have a very different value
in English, for example, from its value in French. Though scholars could
work with some international system of transliteration, the lay reader would
find it quite difficult, and it seems useful to have separate transliteration
systems for the widely used languages, systems that incorporate as many
elements of transcription as possible. Scholars can learn the French or
German transliteration systems, say, in the same way they learn any other
novelty in French or German. And the lay reader will be encouraged to
continue his reading.

A careful writer will generally list the letters he uses for transliteration
in the regular order of the Arabic (or Persian, etc.) alphabet or else refer
to some other publications he follows where such a list is to be found. To
correlate a term in one work with that in another, the reader will need to
refer the letters back to the respective lists; from their position there he
can tell whether they form the same term or a different one. The regular
order is used in the following tables, with the preferable transliterations
indicated and some alternatives added.

Transliteration from Arabic

Alternatives

Inter-

Encyclopaedia Approximate 'Literary'

Sometimes

English

national

of Islam

Pronunciation
glottal stop; as between the
two words in 'me?! angry?'

Used
often omitted

b

English b

t

like English t

th

t

th

English th in thin

t; ts

J

g

dj

English j

g (in Egypt)

h

pharyngeal h ('guttural')

h

kh

h

kh

German and Scots ch,
Spanish j (nearer h than k)

x; k

d

like English d

dh

d

dh

English th in this

d; ds

r

rolled (trilled) r

z

English z

s

s

hissed s (in this)

ss

sh

s

sh

English sh

sch; ch

s

velar s ('emphatic')

ss; s

d

velar d ('emphatic')

dh;d

t

velar t ('emphatic')

th; t

z

velar z ('emphatic')
glottal scrape; to Anglo-
phones difficult to pronounce
sometimes omitted

z

j

10 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

Inter- Encyclopaedia
English national of Islam

gh

q

k

n
h
w

y

a

i
u
a

g

gh

aw
ay

ai

Alternatives

Approximate 'Literary'

Sometimes

Pronunciation

Used

voiced equivalent of kh

g

above

English f

ph

uvular k ('guttural')

c;k

English k

English 1 (in live)

English m

English n

English h

English w

English y (as consonant)

short a as in cat or ask

e

(according to position)

short i as in bit

e

short u as in full

long a as in father; some-

au, o

times as in fat

(but held long)

long i as in machine

ee

(but held long)

long u as in rule

00, ou

(but held long)

English ow in how

ow, ou

English ai in aisle (or in ail)

ey, ei

The transliteration in the table marked 'English' is that usually used in
English-language scholarly publications. 5 In this system some digraphs are

5 The 'English' system is close to that of the Encyclopaedia of Islam except that it
uses j as simpler than dj, and q rather than k because the latter is more expensive and
is confused with k if the dot happens to be lost. The system would have been better if
instead of dots it had used the European accent marks available in every printer's fount
and on every academic typewriter. When such a replacement is made little confusion
results ; and in fact, in some contexts cedillas are regularly substituted for the dots. The
system would also have been better if for ' something else had been used. But in this
work, we follow usual practice, for to introduce a new system without expectation of
getting it generally adopted is merely to add to confusion. In the vowels, either a
macron or a circumflex has been used to indicate length. The circumflex is preferable
as available in ordinary founts and academic typewriters. Scholars might well rethink
the problem of diacritics and their general availability, for we are entering a period of
widely disseminated, inexpensively reproduced works, especially works produced
directly from typescript by offset duplication. (Thus also the old feeling that a given
diacritic should have only a given value universally, which led to using macrons
invariably for length and dots for all velar consonants including q, must yield before
the typewriter and the widening phonetic horizons of our time.) The long vowels might
have been better rendered aa, iy, uw, to avoid confusion on the loss of diacritics, to

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION II

used (e.g., th or sh). In some publications such digraphs are joined by a
bar placed below the pair; when the bar is not used, the rare cases where
the two signs do not form a digraph are indicated by separating them, for
instance, with an apostrophe. 6 (Thus t'h would indicate the pronunciation
of at home.) The most important deviation from these letters still found in
English scholarly works reflects the dislike of digraphs by some purists, who
replace them with the letters listed as 'international' because they are widely
used in Europe. The Encyclopaedia of Islam, edited in more than one
language, has perpetuated a further complication for the two 'English'
choices of j and q. In older works and in journalistic writing the variations
in transliteration are numerous and often based on French phonetic patterns
even though appearing in English language works. The scholarly systems of
other Western languages tend to be similar to the English except for character-
istic variations in the case of digraphs; e.g., English sh is French ch or
German sch; English j is French dj or German dsch.

In addition to the regular consonants, which in the Arabic script appear
uniformly, and the vowels, several other peculiarities of Arabic spelling
must be rendered. Here variations arise from the nature of Arabic writing.
The usual Arabic form as it is written gives less than is needed for sure
identification of a term; for instance it leaves out most 'short' vowels which
all transliteration systems supply. Moreover, transliteration with too great
attention to transcription may confuse the non-specialist reader as to actual
word elements. Any system must admittedly be a compromise, but it seems
better to emphasize the element of reversibility.

The most important points of scholarly compromise are two. One Arabic
consonant is not included in the order of the Arabic alphabet, a certain h
marked like a t, sometimes pronounced h and sometimes t. It occurs only after
an a at the end of a word. In this work it is rendered h after short a, t after
long a; by many it is omitted at least after short a or rendered by t before
the article of a following word. 7 Thus for our Kufah, some write Kufa. The

stress the long pronunciation, and to bring out root letters in the latter two cases; a
nuance of Arabic spelling which distinguishes between u and a hypothetically different
yw is irrelevant to the needs of transliteration. But I follow usual practice.

' Some philologians find a digraph unsound pedagogically. Despite added expense
and the diminished oral recognizability and resistance to accident, they prefer to
multiply diacritics. If persuaded to use digraphs at all, some will use a ligature to make
the digraph artificially a single unit. But typing a ligature when one is also underlining
for italics presents difficulties; moreover, the apostrophe should still be used when the
letters do not form a digraph (such as s'h), to warn the interested but unpracticed lay
reader that normal English pronunciation does not hold.

7 In this case, consistent clarity would call for a preceding diacritically marked vowel,
preferably a specially marked a, but such a letter has not been adopted. The use of
-ah is better than plain -a because it avoids confusion with -a. or -a when the diacritics
are dropped (for instance, distinguishing Hirah from Hira'), a confusion far more
likely than that with an -ah where the h is radical. It also avoids confusion with the
-a of grammatical endings when a whole phrase is transliterated and the endings included.
The use of -at is a concession to widespread pronunciation habits, and overlaps no
more than would -ah.

12 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

second compromise is that the Arabic article al- is in certain cases 'assimilated'
to the noun that follows, the 1 being pronounced like the initial letter of the
noun. 8 Writers attempting to preserve a closer transcription render the
article ar- or as- before r or s, etc., so as to assure a more exact pronunciation;
but many writers, and this work follows such a system, render the article
with an 1, however it is pronounced, because in this case mispronunciation
will cause little confusion and the 1 identifies the article for the layman
(and also answers to the Arabic script form the librarian must learn to
recognize). Thus I write al-Shafi'i, not ash-Shafi'i, which is yet how it is
pronounced.

Minor points that call for notice are the following. Some final vowels
written long in Arabic but pronounced short are here transliterated as long,
but some writers make them short. In the case of a final -a written with a
-y in Arabic, I distinguish it as -a, while some write it as an ordinary long
-a. Consistently with the ordinary rules for vowels and consonants, I write
certain forms -iyy- and -uww-; but some make these -iy-, -uw-. An exception
is usually made for the ending which by normal rules would be -iyy; here I
write -i, to conform with Persian, Turkish, and Urdu, as well as the Arabic
vernaculars, in all which the same ending occurs ; thus I write al-Bukhari, not
al-Bukhariyy. But a few writers make it simply -i; others restore partial
consistency with -iy. Hamzah ('} at the start of a word is always omitted.
Grammatical endings omitted in pausa are always dropped.

Transliteration from Persian, Turkish, and Urdu

In this work I transliterate technical terms throughout in a single form,
usually the Arabic, but the Persian or Turkish when the original form is
Persian or Turkish. Names of persons and countries will naturally vary
from language to language. 9 But in other writers' works even technical
terms, originally Arabic, will often be met with in a Persian form or in still
other forms, including the modern romanized Turkish or an adaptation of it.
For instance, in such works hadith may appear as hadis. Accordingly, the
reader should consider the following notes when reading such works, and he
may find they apply to the Arabic words he encounters as well as to the
other languages.

For the other Islamicate languages transliteration poses certain special
problems. It has been less well systematized by scholars, and this is parti-
cularly true for Persian, from which (and not directly from Arabic) the
Turkish and Urdu and several other languages derived their alphabets.

8 The 'assimilating' letters are d, d, dh, 1, n, r, s, s, sh, t, X, th, z, z.

' As in European languages Maria, Marie, and Mary represent the same name in
different guises, so Husayn, Hoseyn, Hiiseyin represent a single name in Arabic, Persian,
and Turkish. As between Arabic and Persian, at least, the equivalence is easy to trace
by comparing the equivalent letters in transliteration; but for most purposes the
etymology of a name is unimportant to the layman.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 13

Islamicists who are better Arabists than Persianists often treat Persian
words exactly as if they were Arabic, only adding four letters : p, ch, zh, g.
(Zh is pronounced like s in pleasure; for ch, pronounced as in church, c may
be substituted by those who avoid digraphs in Arabic; and for zh, z.) Since
Persian is full of words and names derived from Arabic, and it is often
difficult to decide whether a given name belongs in fact to an Arabic or a
Persian context, such a policy has the advantage of making a decision
unnecessary, and from the point of view of alphabetic reversibility (ignoring
transcription) seems to work well enough.

But Persian is a separate language. Its alphabet has its own rules and is
better rendered by a distinctive system, though preferably differing from
that for Arabic as little as possible. Unfortunately, for transliterating
Persian, especially with a view to at least partial transcription, no one
system has been generally accepted. The following list gives the preferable
systems and some alternatives. {Standard forms for Arabic are shown in
parentheses.)

(th) (dh) (d) (t) (?) (w)

' b p t s j ch h kh dzrzzhsshsztz'ghfqkglmnvhy
s z z t ?

{E.I. :) dj;c [z, z] (E.I.:) k

aiu a i u aw ay ah -i
e o i u ow ey eh -e, -ye
ou, au ei, ai e, a

The first row represents the most widely used system, apart from the
Arabizing system mentioned above. It has the awkward disadvantage (in
addition to diacritics inconvenient on the typewriter, especially when
underlining) of using z for a character different from the Arabic character z
is used for. A newer system, whose points of difference are shown on the
second line, restores t and z to the same use as for Arabic and eliminates the
conflict over z. (Bracketed forms are alternatives to z.) 10

In the newer system, all diacritics (save that for the long a vowel) represent
non-phonetic distinctions — that is, distinctions in Persian spelling that make
no difference in the Persian pronunciation. Hence it fits in with one tendency
in transliterating from Persian which ignores all non-phonetic distinctions
in Persian pronunciation and so, being essentially a sort of transcription,
could almost avoid diacritics altogether. The first line of vowels indicates
transliteration practice close to Arabic; the second line indicates trans-

" This system, adopted by the Library of Congress, seems to carry most authority
at the moment. The Library of Congress has decided to use a bar under the z instead
of the z suggested by its expert and used elsewhere (e.g., by the editors of the Turkish
version of the Encyclopaedia of Islam). The z is wanting in most printing founts and is
apt to be blurred ; but the z is in conflict with usage in the older system as well as being
awkward when a word is underlined to italicize. If a single dot over the z be deemed
inconvenient on the typewriter, it may be easily replaced by an apostrophe or by
diaeresis, as the macron may be replaced by circumflex and the dot under by cedilla.
All these should be available on an academic typewriter.

14 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

literation suggesting a closer transcription. On the third line, are shown
some further variants that are sometimes used.

In this work, I follow the newer system (s, z, etc.) in rendering consonants,
choosing z as clearer than z. But because the vowels of the newer system
have appeared so rarely in scholarly work, and would be badly out of place
in Indo-Persian anyway, I have retained the vowel system that is closer to
the Arabic (using also e, 6 in Indo-Persian), except for the Persian linking
vowel, rendered -e, -ye. The net result is that for Persian words I spell
Arabic-derived names as if in Arabic except for four consonants (w:v; th:s;
dh:z; d:z). Compounds from Arabic are run together (and the article assimi-
lated) rather than separated with hyphens."

All Persian consonants are pronounced nearly as in English, ignoring the
diacritics (which make no difference in Persian pronunciation), except that
kh and gh are as in Arabic, and q is often like gh (it is so transliterated by
some). S is always hissed, g is always hard as in go, zh is like z in azure, w
after kh is silent. Vowels are pronounced as in Arabic, except that long a
tends toward our a in all; short i is rather like e in bed; aw is au in bureau.
Final -ah has the value of e in bed, with h silent. In certain words u. was
formerly 6, and i, e; those words are still so pronounced in India and are so
rendered for Indo-Persian.

Specifically Turkish and especially Urdu names are relatively infrequent
in this work. Modern republican Turkish is naturally to be rendered according
to the modern Turkish Latin alphabet, used since 1928 in the Turkish
Republic (the Turks of the Soviet Union have used other forms of the
Latin and Cyrillic alphabets). There most consonants are pronounced
approximately as in English except that c is our j ; c is our ch; g is standardly
a barely-sounded approximation to Arabic gh; j is the French j (z in azure);
5 is our sh. (S of course is hissed, g is hard.) The vowels are a, e, i, o, u, as
in Italian, with variations; 6, ii, as in German (French eu and u); and i
without a dot (sometimes marked with a circle or a half-circle) — a slightly
stifled vowel found also in Russian.

But the republican alphabet cannot readily be extended to the other
Turkic languages or even to Ottoman Turkish, partly because its characters
conflict with English values and with the system used for Persian and
Arabic, and so would be confusing outside a clearly restricted context.

For the Ottoman Turkish, which used the Arabo-Persian alphabet, our

11 Modern Iranians consistently prefer the e and o, which make the diacritic on !
and ii unnecessary, though perhaps useful where the reader does not know if a name is
Arabic or Persian. But the vowel u is phonetically more accurate in most cases than o,
as is sometimes i than e. Both i and u are usual in scholarly work dealing with pre-
modern periods and occur frequently toward the beginning of words (hence determining
alphabetic position in indexes). It seems justifiable to suggest that e (and ey) and o
(and ow) be substituted for i and u only when diacritics are to be dropped altogether — a
seemingly no longer justifiable practice. As to compounds, they are misleading to
untutored readers if printed separately, and the value of indicating their origin dis-
appears in a second language.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 15

system for Persian will do as a base for the consonants, with the addition of a
form of g pronounced n (originally ng), rendered in this work fi; and replacing
q with k, which eliminates a large class of deviations from modern Latin
and Cyrillic norms. However, pronunciation varied greatly from place to
place and from time to time; in some words g was pronounced y, z and t
were pronounced d, etc. ; final plosives were usually unvoiced. Spelling varied
also. In some of these cases, it is conventional to follow the Istanbul pro-
nunciation rather than the spelling. Often, but not always, the distinction
between k and k, s and s, t and t, is shown in transliteration by the vowel
context ; in such cases, the diacritic may be dropped. The vowels are best
shown by those of the republican Latin alphabet; Arabic aw becomes ev,
ay becomes ey. The three Perso-Arabic long vowels, however, are still
distinguished in quality (not quantity) and can usefully be so marked, but
usually need not be. (The editors of the Turkish version of the Encyclopaedia
of Islam have devised the following system for representing the Ottoman
Turkish :'bptscchhdzrzjsssztz'gfkkgnlmnvhy.)

Other Turkic languages of Persian script are in this work usually assimilated
to Ottoman Turkish or to Persian as seems appropriate, partly because it
is often hard to reconstruct the actual pronunciation of a name in its own
form of Turkish.

Transliteration from Urdu (an Islamicate language of India) cannot be
considered without reference to the received system for Hindi in the Sanskrit
alphabet, for the two are substantially a single language. The older Persian
transliteration (in which Arabic d became z) was developed with the needs
of Urdu in mind and is still the prevailing system for Urdu if there is one.
For Urdu transliteration requires, in addition to the characters for the
Persian letters, vowels in e and 6 and five consonants : four 'retroflex' con-
sonants, written (as for Hindi) t, d, r (all different from anything in Arabic),
and a sign for vowel nasalization, n (or n or tilda over the vowel).
Many consonants in Urdu appear in an aspirated form, that is, accompanied
by an h that represents a puff of breath after the stop and not the formation
of a fricative. Thus th must be read as a strong t, and so for dh, th, dh, ph,
bh, kh, gh, chh, jh, rh. Hence kh and gh, as digraphs, can within the Indie
context come to stand for two different consonants each: the fricative kh
and gh as in Persian and Arabic, and the aspirated kh and gh. (Duplication
for th and dh does not occur in that context.) The digraphs can be joined
by a bar beneath them when they form a fricative, and not when they
form an aspirate. (For Hindi, English ch is often written plain c, and aspirated
ch, sometimes transliterated chh, is written ch.)

For the purposes of this work, it seems simpler to transliterate Urdu,
when necessary, by the same system as used for Persian, only adding the
retroflex t, d, f (as in the Encyclopaedia of Islam), n (and nasal h). (The
fricative kh and gh may be so underlined ; but this merely distinguishes the
one digraph from the other, for aspirate kh and gh are equally digraphs and

l6 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

must not be confused with what could be marked k'h, g'h discussed in the
section on Arabic transliteration.) Except for retrofiexes, aspirates, and
nasalization, pronunciation of consonants is much as in Persian (commonest
English sound, with diacritics ignored). Vowels are as in Arabic except: a
like u in cut, ay like a in bad, aw as in awe, e as in they, 6 as in over.

As to other languages, there is little occasion to transliterate from them
in this work and when occasion arises it will be done, so far as possible, on
one of the following bases. Sometimes there is an agreed scholarly convention
which can be followed. When there is a standard modern romanized form,
that will do even for earlier times; or the transliteration may follow a
Cyrillic script. Sometimes names will be assimilated to one of the above
four languages, when they appear in such a context. The transliteration of
the four languages into English in this work is therefore (for the reasons
given in text and notes) :

Arabic : 'b t thj hkhd dhr z sshsdtz'ghfqk lmn why

Persian : 'bpt sjchhkhd z r zzhsshsztz'ghfqkg lmn vhy

Urdu : 'bptf sjchhkhddzrfzzhsshsztz'ghfqkg lmnn vhy

Ottoman

Turkish : 'bpt sjchhkhd 2 r zzhsshsztz'ghfkkgfllmn vhy

Arabic : a i u a i u aw ay -ah

Persian : a i u a 1 u aw ay -ah -e, -ye

Urdu : aiuaSiou aw ay -ah

Ottoman

Turkish : (aeiouoiiiaiu)

By such systems, transliteration for all the languages is fully reversible
(except some details of Turkish); does not conflict with that for Arabic
(except in the Urdu aspirates th, dh) ; allows for a fair degree of oral recog-
nizability; offers a minimum of difficulty in being adapted to typewriters
and printers' founts ; and suffers minimal loss of recognizability if diacritics
are dropped. The systems used in this work deviate from the second edition
of the Encyclopaedia of Islam as follows :

in all languages: -ah for -a, q for k (except Turkish), j for dj ; no digraph

bars (unless in Urdu) ;
in Persian, Urdu, Turkish: s for th, z for dh, i for d, v for w, ch for c,

-e(-ye) for -i, and assimilation of article and compounds {-uddin etc.) ;
in Indo-Persian and Urdu also: e, 6;
in Turkish from Arabic also : the Turkish vowels.

MUSLIM PERSONAL NAMES

Many of the personal names to be met with in Islamicate materials will
seem long and forbidding to the beginner and hard to sort out. This is
natural in the case of an alien civilization; the only complete remedy for the
difficulty is long familiarity. Nevertheless, the reader can take note of

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 17

certain classes of personal names which he can learn to identify. If he can
recognize what is similar in these names, he can then more easily concentrate
on their distinguishing features when he meets them in his reading, and so
keep them apart.

Muslim personal names have been mostly of Arabic, Persian, or Turkish
origin, even in countries with quite different languages; the Arabic element
is especially strong. Other languages are of course locally represented, as in
Malaysia or Africa. The basic approach to personal naming has been by
and large the same everywhere. Until modern times very few people in
Muslim countries had 'family' names. The commonest way of naming was
threefold: first, a 'given' name, received at birth; then the given name of the
father; and finally, if necessary, one or more descriptive names, such as one
indicating the man's city of origin or his occupation or perhaps his ancestry,
his family. In addition, a man of position commonly had one or more honorific
names, often bestowed by a ruler. In several times and places there have
been, to all intents and purposes, two given names: a specifically Muslim
name, usually of Arabic derivation, alongside one of local derivation; an
ordinary name alongside one honorific in form ; or, among Arabs, a kunyah
(explained below) alongside an ordinary name. Usually only one or the
other of these was much used.

No one of these various names has necessarily been the 'filing name' — the
one by which one would look up the man in a card catalogue or in a telephone
directory. What part of the total appellation is to be used for such quick
reference has been a matter of convenience — sometimes whichever of a man's
names was the least common among his fellows. For instance, suppose a
man's given name is Ahmad, his father's name is 'Alt, and he is known
as Zinjani because he comes from Zinjan. There are dozens of Ahmads in
any Muslim city, and dozens of 'Alis; but very few men from the small
place of Zinjan — so the least ambiguous way to refer to the man (except
as regards other men from the same place) is Zinjdni. Often there have been
two or more such names by which a man was known, perhaps in different
circles. Accordingly, in referring to a man it is best for the beginner to
follow the lead of whatever author he is depending on, and in cases of doubt
to use a combination of as many names as necessary.

There are several elements, found especially in names of Arabic origin,
which form an inseparable part of certain sorts of names, and which the
student should learn to recognize. Al- is simply the definite article, 'the',
and occurs with a great many names. Before certain consonants the I is
assimilated to the consonant, being pronounced — and in some transliterations
written — r, s, d, etc. (Cf. the section on Transliteration.) In some cases
the al- can be either present or absent, indifferently. It is usually omitted
in indexing. Abu originally meant 'father of, but has come to be, in effect,
merely an element in a man's name: any given name can be added to it,
the combination forming a single new name. Among the ancient Arabs

l8 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

this was used as a sort of epithet or honorific, the kunyah, in addition to the
more personal name, e.g., Abu-Bakr. [Abu- in Persian and Turkish some-
times becomes Bit- or B&-.) Umtn- has the place in women's names of
Abu in men's, e.g., Umm-Kulthum.

The letter b is short for ibn, meaning 'son of (or for bint, 'daughter of)
which is often used in Arabic between one's given name and the father's
name — and further between the father's and grandfather's names, and the
grandfather's and great-grandfather's, and so on up the line, which is
sometimes traced very far in giving a man's full appellation. Tbn-N . . .'
sometimes becomes simply a family name. In languages other than Arabic,
and in many modern Arabic lands, the ibn is usually not used. On the other
hand, in Persian its place may be taken by the suffix -zddah; in Turkish
by the suffix oglu, which commonly form family names. Sometimes these
are interchanged; thus Ibn-Taqi = Taqi-zadeh.

N.B. : Such elements as Ibn- or Abu-, as well as certain other prefixed
(or suffixed) words like 'Aid-, cannot be dropped from a name without
basically changing it. In this work they will be hyphenated. 12

Classes of names to be spotted and kept distinct (this is in no sense an
exhaustive list of types of names — it includes only easily recognizable
types) :

I. Names of prophets as given names:

Most of the Biblical heroes are recognized as prophets by the Muslims, and
their names, in Arabic form (sometimes further modified in other languages),
are commonly used. (Christians and Jews in Islamicate lands have used
special variants of some of these names.) Examples are : Ibrahim = Abraham ;
Isma'il = Ishmael; Ishaq = Isaac; Ya'qub = Jacob; Yusuf = Joseph;
Musa = Moses; Harun = Aaron; Da'ud = David; Sulayman = Solomon;
Yahya = John; Maryam = Mary; Tsa = Jesus. In addition, there are
several prophets not mentioned in the Bible, notably Muhammad himself,
whose name is very commonly used.

II. Names from favoured Arabic roots: notably

(a) in HMD, three consonants which enter the formation of words having
in common some notion of 'praise', thus Muhammad, Ahmad,
Mahmud, Hamid, etc.;

12 The use of the hyphen in compound proper names is called for because of the
Anglo-Saxon habit of isolating the last element in any personal appellation as a primary,
'last name'. (The use of the hyphen can help to eliminate a remarkably prolific source
of confusion in beginners.) Subordinate and inseparable elements in names include:
Ibn-, Bint-, Vmm-, Ab-A-, Ntir-, Gholdm-, MamMk-, 'Abd-, Dh-U-, Sibt-; and of suffixed
words: -qoli, -bandeh, -bakhsh, -dad, -zddeh, -oglu. Many other compounds are similarly
used as names. In some cases, the subordinate part of the name can be dropped almost
at will or else may be used alone; but the hyphen is appropriate wherever the last
element in the name cannot properly be used by itself ; the reader should not overlook it.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 19

(b) in HSN, forming words having in common some notion of 'good',
thus Muhsin, Hasan, Husayn, Hassan, etc.;

(c) in S'D, forming words having in common some notion of 'happiness',
thus Sa'd, Sa'id, Mas'ud, etc. ;

(d) in ZYD, forming words having in common some notion of 'growth',
thus Zayd, Yazid (Yaziyd), etc. ;

(e) in 'MR, forming words having in common some notion of 'life', thus
'Amr, 'Umar, etc.

III. Names in 'AM:

The word 'abd means 'slave of, and is prefixed to epithets of God, to form
single, indivisible given names. Thus, 'Abd-al-Qadir, 'Slave of the Almighty'.
(Hence such names cannot be split up, as sometimes happens in the press.)
Other common names of this class are 'Abd-al-Rahman (or Abdurrahman,
in another spelling), and the common 'Abd-Allah (more usually, 'Abdullah).
Other words meaning 'slave' are used in the same way: thus the prefixes
Mamluk- and Gholdm- and the suffixes -bandeh and -qoli. Often these are
attached to the name of a prophet or an imam rather than of God.

IV. Names in -Allah:

Analogous to the above class of names are those ending in -Allah (or some-
times other divine names) with various first elements (or, sometimes, suffixes) ;
thus, Hamd-Allah, 'Praise-God'. Sometimes similar compound names are
formed with a prophetic name; thus, Nur-Muhammad, 'Light of Muhammad'.

V. Names in -al-dtn, -al-dawlah, -al-mulk :

These were originally titles but have often become simply given names. For
example, -al-din (or -eddin) means 'of the religion': thus, Qutb-al-din,
'Pole of the religion' ; -al-dawlah (or -eddaula) means 'of the dynasty' : thus,
Mu'izz-al-dawlah, 'Strengthener of the dynasty'; -al-mulk means 'of the
kingdom': thus, Nizam-al-mulk, 'good order of the kingdom'. N.B. : These
three post-fixed elements sometimes can be dropped without deforming the
name beyond recognition.

VI. Names in Mm-:

A majority of Arabic participles start with M, and especially with Mu; and
a great many names, originally often honorifics, are in participle form. This
is true of the reign names of most of the 'Abbasid caliphs : Mansur, Ma'mun,
Mu'tasim, Mutawakkil, Mu'tadid, Mustansir. Often these start with Mut-
or Must-, which form participles in whole sets of verbs. It is the letters
following the initial mu-, that must be taken note of in order to keep names
of this class apart.

20 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

VII. Names in -i:

This is the common termination, taken over into several Islamicate languages,
indicating origin or relationship. A person from the town Shiraz is SMrdzi;
a person from India (Hind) is Hindi; a protege of a man called Sa'd is
Sa'di; a descendant of 'Uthman is 'Uthmdni; a member of the tribe of
Kindah is al-Kindi. (The Turkish equivalent ending is -li or li, -lu, -hi.)

Constantly repeated, but not easy to classify by sight, are various other
personal names of Arabic origin (masculine: 'Ali, Ja'far, Habib, Hamzah,
Salih; feminine: Fatimah, Zaynab, etc.), of Persian origin (Firuz, Bahrain,
etc.), of Turkish origin (Arslan, Timur, etc.), and of still other origins.

Sometimes various titles are used as parts of purely personal names — for
instance Malikshah, in which malik is Arabic and shah Persian for 'king'.
More common is the use of titles or titular elements to replace a personal
name, or used inseparably with it. Among such titles are mir {amir), beg,
khan, mirzd, shaykh, shah, dghd, (dqd), etc. Some titles have a special signi-
ficance: shah as prefix often means 'saint', as suffix, 'king'; mirzd as suffix
means 'prince', as prefix, 'sir'; sayyid or shartf is used for a descendant of
Muhammad; hdjj, or hdjji, for one who has made a pilgrimage, not necessarily
to Mecca.

In various languages, the same name can sometimes be scarcely identifiable :
thus Muhammad becomes Mehmed (or Mehmet) in Turkish, Mahmadu in
West Africa. In the Russian empire, Muslims themselves have commonly
replaced -i, -oghlu, -zddeh by -ev or ov; or simply added the latter (Russian)
endings to names as they stand. Muslims in India have often Anglicized
their names in ways answering to vague norms of English pronunciation;
thus Syed for Sayyid, Saeed for Sa'id; more generally, ay = y, i = ee,
u = oo, and short a becomes u. In French colonies, Muslims Gallicized their
names: s = ss, u = ou, g = gu, etc.

THE ISLAMICATE CALENDARS

An era is a system of numbering years from a given base year — thus the
year i of the common Christian era is the year in which Christ was sup-
posedly born. The year i of the Islamic era is that of the Hijrah (H.), in
which Muhammad moved from Mecca to Medina and effectively established
the Muslim community. That year corresponds to the last part of 622 and
the first part of 623 of the Christian, or Common, era (ce) .

In contrast to the case with most eras, the 'years' of the Islamic era are
not true solar years — complete rounds of the seasons — but are lunar 'years'
of twelve 'true months', twelve periods from new moon to new moon. Since
a true month has only 29 or 30 days, twelve such months fall short of a
solar year by about eleven days. Most lunar calendars that have used true
months have kept themselves in line with the solar year by adding an extra

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 21

month every three years or so. This was forbidden in the Qur'an, and in
consequence a date in the Islamic 'year' has no fixed relation to the seasons;
a given festival, for instance, will pass from summer to spring to winter
and autumn and back to summer again three times within a long lifetime.
Accordingly the Islamic calendar, while it has been used for ritual and
historical purposes, has almost always and everywhere been accompanied
by a different, a 'secular' calendar, one of solar years, which could be used
for fiscal and other practical purposes. These secular Islamicate calendars
have been numerous but have never had much prestige, and few of them
have been in use long enough, or reckoned systematically enough, to serve
for historical purposes. Except in recent times, when the secular calendar
used has usually been the Gregorian or modern international calendar, no
such auxiliary solar calendar has seriously rivalled the Islamic lunar calendar
in general acceptance.

Because the Islamic era does not reckon by solar years even approximately,
it is not possible, as it is in the case of other eras, simply to add to the Islamic
date the difference in years between the two starting points (622), in order
to get the date in the Christian era. If one does not have a conversion table,
there is a quick method of finding the date to within a couple of years.
Because of the shortness of Islamic 'years', an Islamic century is accomplished
in three years less than a century of solar years; hence the Islamic dates
are always gaining on the Christian at the rate of three in a hundred years.
At year 1 of the Hijrah there was a difference of six centuries plus 21 years —
Christian year 622. At year 100 there was a difference of six centuries plus
18 years (100 + 618 = Christian year 718) ; at year 200, of six centuries
plus only 15 years (815 ce) . By the year 700 the difference is just six centuries
— Christian year 1300. After that, it is a difference of six centuries minus
so many years. Accordingly, the approximate date in the Christian era can
be found by adding to the Islamic date 600 years plus three years for every
century before 700 (H.), or minus three years for every century after 700 (H.).

For the arithmetically inclined, a somewhat more exact date can be

TT

gained with the formula G = H — — + 622 (G = Gregorian date, H = Hijrah

33
date). 13 The Islamic date of a Gregorian year can be found with H = G — 622

G — 622

-| . It must be remembered that such formulae give only the year

32

during which the corresponding year began — the greater part of the latter

year may have coincided with a year succeeding the one given.

The months of the Islamic year traditionally have been determined by

actual observation of the new moon. Hence the same month had different

numbers of days in different years and even in the same year in different

13 Somewhat easier to use might be the formula stated in the following terms:
G = H — .031 H + 622.

22 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

places; this makes exact dating precarious unless the week-day is also
known (for the Islamic calendar included the seven day week as well as
the month). Under some regimes mathematical means, such as alternating
the months thirty, twenty-nine, thirty, and adding 'leap years' by various
systems, were used to make the calendar more predictable. The months are
in order: Muharram, Safar, Rabi' I, Rabi' II, Jumada I, Jumada II, Rajab,
Sha'ban, Ramadan, Shawwal, Dhu-1-Qa'dah, Dhu-1-Hijjah.

Since the month began with the sighting of the new moon in the evening,
the day was made to begin at sunset.

The most common month names of the solar Islamicate calendars are
(with appropriate Gregorian equivalents)

for Arabic:

Nisdn (April), Ayydr (May), Hazirdn (June), Tamuz (July), Ab (August),
Ayhll (September), Tishrin al-awwal (October), Tishrin al-thdni (November),
Kdnun al-awwal (December), Kdnun al-thdni (January), Shubdt (February),
Adhdr (March) ;

for Persian:

Farvardin (March- April), Urdi-bihisht (April-May), Khurddd (May-June),
Tir (June- July), Murddd (July-August), Shahrivar or Shahrir (Aug.-Sept.),
Mihr (Sept.-Oct.), Abdn (Oct.-Nov.), Azar (Nov.-Dec), Day (Dec.-Jan.),
Bahman (Jan.-Feb.), Is/and (Feb.-March).

HISTORICAL METHOD IN CIVILIZATION STUDIES

Historical humanism

Unless a scholar is content to accept his categories (and hence the questions
he can ask and hence the answers he can arrive at) as given by the accidents
of current predispositions, he cannot escape the obligation of justifying his
selection of units for study, which means justifying his point of view. Such a
justification, in turn, must imply an explicit stand on his role as a scholar.
If there were unanimity in these matters, they might be left tacit — at least,
if the given scholar were in accord with the rest. Fortunately, several quite
different viewpoints guide historical studies generally, and Islamic studies
in particular, in our present world.

Historical studies have been called 'idiographic' as describing dated and
placed particulars, as do many phases of geology or astronomy, in contrast
to 'nomothetic' studies such as physics and chemistry, which are supposed
to lay down rules to hold regardless of date. This distinction has its use-
fulness so long as one bears in mind certain considerations sometimes
forgotten. Firstly, whether the objects of the questions are dated or dateless,
the questions themselves (as befits a cumulative public discipline) ought to
be, in some degree, of timeless significance to human beings: sometimes
perhaps leading to manipulative power, but always leading to better under-

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 23

standing of things that matter to us humanly." Moreover, any discipline,
ideally, should not be defined exactly by the category of the objects it
studies nor even by the methods it uses, and still less by the form of its
results — though empirically these may be useful indices, especially in
interpreting the various academically recognized fields of inquiry which
have grown up largely by historical accident. Ideally, a discipline needs to
be set off just to the degree that there is a body of interdependent questions
that can be discussed in relative autonomy from other bodies of questions,
at least according to some one perspective. In a discipline so set off, it
cannot necessarily be decided in advance just what forms of questions will
prove to be required or what sorts of methods will prove necessary to answer
them effectively. From this point of view, if there is a field of historical
studies (as I believe) and not merely a group of several fields, it can be
nothing less than the whole body of questions about human cultural develop-
ment, about human culture in its continuity over time ; and here we cannot
rule out a potential need to develop relatively dateless generalizations, for
instance about what may be possible in cultural change, such generalizations
are not simply derivable from any other discipline as such, yet they are
necessary for studying what is timelessly important about the dated and
placed events of human culture.

These considerations being understood, then it can be said that historical
studies of human culture are preponderantly 'idiographic' in the sense that
even their broader generalizations are usually not dateless, in contrast to
certain kinds of nature study, and perhaps in contrast also to certain kinds
of social studies of human culture, designed to refine analysis of any given
society at any given time. Moreover, in any case, historians' questions are
concerned ultimately with the dated and placed, and when (as they must)
they ask questions that are undatable within the historical context, it is
for the sake of elucidating particulars which are, dated and placed, however
broad in scope, and not vice versa. The dated and placed events are not
mere examples, not mere raw material for dateless generalizations.

But I am concerned here with a further distinction. Within the body of
questions about culture in its continuity over time, even when the focus of
interest is admittedly on the dated and placed as such, one can still dis-
tinguish historical viewpoints further in terms of what sort of date-bound
questions are regarded as primary, the answer to which is the goal of the
inquiry; and what sort are regarded as subordinate, yielding information
which will help in answering the primary questions. On this basis we may

" Increased predictability through the 'lessons' of history, and hence increased power
of manipulation, may sometimes supervene through historical study; but it is surely
not its true purpose. On the other hand, prediction as a means of verification sometimes
plays an essential role in historical inquiry. This is not, of course, prediction of 'the
future' — that is not the proper purpose of any scholarly or scientific discipline — but
prediction of future evidence, which may come in the form of laboratory experiments,
of field surveys, or (in the case of history) of newly found documents.

24 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

distinguish two sorts of historian, 'typicalizers' and 'exceptionalizers'. In
practice, the distinction is one of emphasis: the 'exceptionalizer' is concerned
with all that concerns the 'typicalizer', or he ought to be; and despite his
principles, the 'typicalizer' generally finds himself involved in points he
might feel should concern the 'exceptionalizer' alone. Nonetheless, the two
viewpoints can issue in the use of differing units and categories in defining
the field of study. I believe that in pre-Modern civilization studies, at any
rate, the more inclusive view that I am labelling 'exceptionalizing' cannot
be left out of account if the humanly most significant questions are to be
got at. It is on this principle that I have constructed this work.

Some historians, relatively 'typicalizing' in viewpoint, intend primarily to
articulate intelligibly their chosen portion of the total cultural environment
as it impinges by way of interacting events on the present human inquirer.
They intend to present that environment as it is structured in space and time
(asking, in effect, how things came to be as they are now) much as an
astronomer studies the particular structure, in space and time, of the solar
system. Some may even hope that their work may ultimately serve chiefly
to elucidate dateless regularities of culture change, not tied to any dating
or placing (at least within the particular span of time and place which
human culture as a whole presupposes). Such historians, if fully consistent,
must be concerned first with the typical, and then with the exceptional only
as it serves to make clear (or perhaps account for) what is or has been
typical. If they study a state, or a novel, or a sect, they will study it primarily
as typifying, or at least causing, general political or aesthetic or religious
patterns — at least the patterns of the time, and perhaps preferably those of
all time.

On the other hand, from what may be called a more humanistic viewpoint,
the reason for studying the typical is rather that thereby we may be better
able to appreciate the exceptional, seeing more fully in just what way it is
exceptional. We need to know works of artists or acts of statesmanship
which are typical of a period j ust so that we may the better place the excellent,
the outstanding. 15 We study Islamdom as a whole, as a great complex historic
event, as well as the various less extensive events that compose it, not
primarily as examples of something more general but as something un-
recurrent and unrepeatable, and as having importance precisely for that
reason. In consequence, we can be as concerned with the great failures as
with the great successes, and as concerned with the potential moral implica-
tions of an act as with its immediate outcome.

Such inquiry remains legitimate public inquiry, and not just private

15 This is not reducible to aesthetic criticism, of course, let alone to straight moral
judgment. The difference between the art historian and the art critic — and the corres-
ponding difference in other fields than art — is a matter of the historian's concern with
culture as such in its dimension of continuity over time. But such a concern cannot
do away with the sense of greatness; it rather puts it in perspective.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 25

antiquarianism, to the extent that the exceptional events were in some
sense or other outstanding in the context of mankind generally, and not
just for private individuals or groups. Events evidently meet this test when
they have altered the context of routine human life in their time, insofar as
no region or period of human life has, in the long run, been so isolated that
it has not had its effects in turn on the rest of us. On this level, the 'exception-
alizer' is at one with the 'typicalizer'. But he wants to add a further dimension.

It is not merely as events have altered the natural or the socio-cultural
context that they can have exceptional significance. So far as there is moral
or spiritual solidarity among human beings, apart from physical con-
frontations at any given time, the fate of each people is relevant to all human
beings whether or not it had permanent external consequences otherwise. It
is, then, also, and perhaps above all, as events and acts have altered the
moral context of human life that they are of universal significance, for they
have set irreplaceable standards and norms, and they have posed distinctive
challenges and established moral claims which as human beings we dare
not ignore. Herodotus wrote his history, he said, to preserve the memory
of the great deeds done by the Greeks and the Persians: unrepeatable deeds
that have an enduring claim to our respect. Those deeds cannot be imitated,
though they may be emulated and in some sense perhaps surpassed. But
even now we dare call no man great whose deeds cannot somehow measure
up to theirs. Once having known those deeds, the world can never be quite
the same for us again: not because of what they may tell us of what we are,
may tell us statistically about the potentialities of our hominid species; but
because they add to our understanding of who we are, of what we are
committed to, as human beings, what is worthy of our wonder and our tears.

We are speaking here of such events and acts as form human cultural
institutions on the level of public action. We are dealing with peoples — or,
more accurately, with groups of men and women at least relatively auto-
nomous in culture. Purely individual exploits may have something of the
same quality, but they are meaningful on a different level and their student
is the biographer, not the historian. Yet it is especially in this 'exception-
alizing' perspective that persons' ideal norms and expectations and even
the special visions of individuals can be crucial. For they prove to be the
mainsprings of creativity at the interstices of routine patterns, when
exceptional circumstances arise and something new must be found to do.
This is how, in fact, the would-be 'typicalizer' finds himself dragged into
matters more suited to the programme of the 'exceptionalizer'.

Clearly, the serious 'exceptionalizer' — despite the doubtful example of
some scholarly story-tellers — necessarily needs to understand all that the
most 'social-scientific' of the 'typicalizers' will want to be studying. Always,
of course, visions and ideals can come into play only within the leeway allowed
by the human interests (material and imaginative) of those less concerned
with ideals. Ultimately all historical 'why's' must be driven back (often in

26 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

the form of 'how could that have become effective?') to circumstances of
hominid natural and cultural ecology — the circumstances which determine
that what would otherwise be the individual random 'accidents' that shape
history will not simply cancel each other out but will be reinforced and
cumulatively lead in a single direction, 16 However irrational human beings
may be, in the long run their irrationalities are mostly random. It is their
rational calculations that can be reinforced in continuing human groups
and can show persisting orientation and development — even when they are
calculations on misconceived presuppositions. 17 Hence group interests have a
way of asserting themselves. Group interests seem ultimately based in
ecological circumstances in general and, more particularly, in that cumulative
development of cultural resources which the essential internal instability of
cultural traditions assures will be likely, in the long run, to be ever more
elaborated and so to require ever new adjustments.

But such ecological circumstances merely set the limits of what is
possible. Within those limits, the personal vision has its opportunity. For
when habitual, routine thinking will no longer work, it is the man or woman
with imagination who will produce the new alternatives. At this point, the
concerned conscience can come into play. It may or may not prove adequate
to the challenge. But in either case, it is such personal vision that is the most
human part of human history.

Hence the humanistic historian must concern himself with the great
commitments and loyalties that human beings have borne, within which
every sort of norm and ideal has been made explicit; and he must concern
himself with the interactions and dialogues in which these commitments
have been expressed. Hence, for an 'exceptionalizing' historian with such
intentions, it is Islamdom as a morally, humanly relevant complex of
traditions, unique and irreversible, that can form his canvas. Whether it
'led to' anything evident in Modern times must be less important than the
quality of its excellence as a vital human response and an irreplaceable
human endeavour. In this capacity, it would challenge our human respect
and recognition even if it had played a far less great role than, in fact, it
did play in articulating the human cultural nexus in time and space and
in producing the world as we find it now.

On scholarly ^recommitments

Because of the central role, in historical studies, of human loyalties and
commitments, the personal commitments of scholars play an even greater

16 It is for this reason that every 'why this?' presupposes at least one 'why not that?'
'Might-have-beens' are built into the inquiry of any historian, whether explicit or not,
just as they are built into that of any other scholar or scientist.

" On the self-determination of each new generation — as against 'blind tradition' —
compare the section on determinacy in traditions, below.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 27

role in historical studies than in other studies, a role that stands out in special
relief in Islamics studies.

On the most serious levels of historical scholarship — where the human
relevance of major cultural traditions is at issue, such as that of religious
or artistic or legal or governmental traditions, or even that of whole civiliza-
tions — historical judgment cannot be entirely disengaged from the basic
precommitments of inquirers. Indeed, it is not necessarily desirable that it
should be : the very issues can arise only as we are humanly deeply engaged.
Inquiries by pure specialists, seeking only to straighten out this or that
detail brought up by some greater scholar who was humanly engaged and
had discussed the great issues, may bring useful clarifications but often miss
the main points. Precommitment can lead the unwary — and often even the
most cautious scholar — to biased judgment. Bias comes especially in the
questions he poses and in the type of category he uses, where, indeed, bias
is especially hard to track down because it is hard to suspect the very terms
one uses, which seem so innocently neutral. Nevertheless, the bias produced
by precommitment can be guarded against ; the answer to it cannot finally
be to divest ourselves of all commitments, but to learn to profit by the
concern and insight they permit, while avoiding their pitfalls.

Such basic precommitments are always to a degree idiosyncratic in really
serious scholars; yet the deeper they are, the more fully they are likely to
be rooted in one of the major cultural traditions of ultimate overall com-
mitment. In fact, certain of these traditions have loomed especially large in
determining the viewpoints of the masters of Islamics studies, who have
done the most to set the problems and the framework within which other
Islamicists have worked. I shall mention five, three old and two new. The
Christian tradition — in Catholic or in Protestant form — has been deeply
determinative for many Western scholars, as has Judaism for still others.
More recently, increasing numbers of scholars committed to the Islamic
tradition — Shari'ah-minded or Sufi — are making their contributions to
scholarship in the field. The pitfalls that await scholars committed to any
of these traditions are evident enough in such scholars' work, at least to any
scholar of a rival commitment. It is no guarantee of balanced insight, to be
a Muslim, nor of impartiality, to be a non-Muslim. Alongside these older
traditions, and representing precommitments leading to the same sorts of
pitfalls as lurk in commitment to Christianity or Islam, we find Marxists
on the one hand and dedicated Westernists on the other. I call 'Westernists'
those whose highest allegiance is to what they call Western culture, as the
unique or at least the most adequate embodiment of transcendent ideals of
liberty and truth. They usually share, to some degree, a Christian viewpoint
on Islam, insofar as the Christian tradition has been so central to Western
culture, however much personally they may reject the claims to allegiance of
Christianity in itself. Not all Islamicists are consciously committed to one
of these major allegiances; but for many who are not, the alternative is not

28 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

genuine independence and objectivity. Commonly the alternative, rather, is
more limited horizons and shallower awareness, together with unconscious
and hence unanalyzed piecemeal commitment to partisan viewpoints which,
in those consciously committed, are subject to conscious review and
control.

Accordingly, the problem of how one may legitimately go about studying
Islam from within a commitment to another great tradition — and in parti-
cular how to go about studying it from within a Christian commitment — is
no by-problem of interest only to a few scholars who by exception are
religiously inclined. It is central to the whole scholarly problem. Jean- Jacques
Waardenburg, in L' Islam dans le miroir de V Occident: comment quelques
orientalistes occidentaux se sont penchis sur VIslam et se sont formes une
image de cette religion (The Hague, 1963), has demonstrated how the work
of the formative Islamicists Ignaz Goldziher, Christiaan Snouck Hurgronje,
Carl Becker, Duncan MacDonald, and Louis Massignon was in each case
intimately and pervasively marked by the basic precommitments of these
men (though he does not use the concept 'Westernist'). The cultural allegiance
of the serious scholar is crucial in his work. This is not to say that it is
impossible to study fairly one religious tradition from within another, as
has sometimes been suggested. Ultimately all faith is private, and it is
often far easier for congenial temperaments to understand each other across
the lines of religious or cultural tradition than it is for contrasting tempera-
ments to make sense of each other's faith even when they follow the same
cult and utter the same creed. We are primarily human beings and only
secondarily participants in this or that tradition. Nevertheless, not only the
scholars' cultural environment at large but their explicit precommitments,
which brought the greater of the scholars to their inquiry in the first place,
have determined the categories with which they have undertaken their
studies. Only by a conscious and well-examined understanding of the limits
of these precommitments and of what is possible within and beyond them
can we hope to take advantage of our immediate humanness to reach any
direct appreciation of major cultural traditions we do not share — and perhaps
even of traditions we do share.

Where we compare the Occident and Islamdom in general, and Christianity
and Islam in particular, such awareness is especially essential. There has
been a tendency, among those Christians who have been willing to concede
spiritual validity to Islam at all, to see Islam as, in one way or other, a
truncated version of Christian truth: all or virtually all the truth to be
found in Islam is to be found in Christianity, but Christianity leads beyond
that truth to a crowning essential truth that eludes the Muslim's grasp.
Correspondingly, Muslims have historically seen Christianity as a truncated
or perverted Islam. But such a comparison is, on the face of it, unsound at
least for historical purposes. It can hardly be intelligible, to those Christians
or Muslims having such views, how it can be that intelligent, sensitive, and

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 29

upright persons can prefer Islam to Christianity, or vice versa, once they
have heen exposed to the appeal of both.

In sensitive hands, some such approach can have suggestive results,
indeed. The most attractive such interpretation of Islam from the Christian
side is surely that of Louis Massignon, set forth allusively in a number of
his articles, such as 'Salman Pak et les premices spirituelles de l'lslam
iranien', Societe des Etudes Iraniennes, vol. 7 (1934), and in his several
articles on the Seven Sleepers; he saw Islam as a community in spiritual
exile, veiled from the divine presence, yet through that very exile charged
with a special witness to bear. (Giulio Basetti-Sani, Mohammed et Saint
Frangois [Commissariat de Terre-Sainte, Ottawa, 1959], has developed part
of Massignon's idea in his beautiful and knowledgeable, if not very scholarly,
book, which forms a suggestive contribution to a modern mythology.) A
less poetic, though still sensitive, approach to Islam in Christian terms is
offered by Eric Bethmann's Bridge to Islam (Nashville, Tenn., 1950) and
by the works of Kenneth Cragg. Yet it remains true that the ultimate
judgments such approaches presuppose are suspect. A serious exploration of
any one religious tradition in its several dimensions could consume more
than one lifetime, and it is not to be expected that many persons can
genuinely explore two. If this fact helps account for so many intelligent
persons not seeing the truth as the apologist sees it, it also suggests that the
apologist too is deceiving himself if he thinks he is qualified to judge the
rival tradition. A view of Islam as a Christianity manque, or the reverse,
however elegantly formulated, must be received with great scepticism.

But the readiest alternatives, among those willing to concede some truth
to a rival tradition, are equally unsatisfactory for making a comparison.
One may resort to syncretistic assimilation, as if superficially similar elements
in the two traditions could be identified; but this is bound to falsify one
tradition or both — if only by not recognizing the genuineness of the demand,
at the heart of each, for exclusive historical commitment. For instance, in
both traditions there is a demand for moral behaviour on the basis not of
arbitrary human custom but of divine revelation; and at least in broad
areas, the moral norms implied in the two revelations are much alike. Yet
for Christians, being based in revelation means being in response to redemptive
love as it is confronted through the presence of a divine-human life and
the sacramental fellowship of which that is the source. For Muslims, being
based in revelation means being in response to total moral challenge as it
is confronted in an explicit divine message handed on through a loyal
human community. The two senses of revelation not only contrast to one
another: they exclude one another categorically. Yet to abstract from them
is to make pointless both the Christian and the Muslim demand for a revealed
morality over against human custom.

To avoid the over-explicit identifications of syncretism, one may resort
frankly to reducing both traditions to some lowest common denominator — a

30 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

formless mysticism or a vague appeal to the common goodwill of mankind.
But in practice this means appealing to the prestige carried by the great
traditions, on behalf of something that can rise above the level of impotent
platitudes only as the quite private viewpoint of an individual.

The two traditions, as such, must be recognized as incompatible in their
demands, short of some genuinely higher synthesis presumably not yet
available to us. And we must retain this sense of tension between them
without interpreting the one by the standards of the other. This may be
accomplished, in some degree, through a comparison of the two structures,
of what sorts of elements tend to get subordinated and what tend to get
highlighted. In such a perception, those committed and those with no
commitment can join, provided each maintains a sensitive human awareness
of what can be humanly at stake at every point. But this is possible only so
far as the elements chosen can be evaluated in some independence. This is
an ideal only approximable at best. Hence even the best comparison cannot
be regarded as providing an objective basis for ultimate judging between
traditions. Yet it may make more understandable the special strengths of
Islam — and its weaknesses — in the given historical circumstances.

I have developed this point about the irreducible incompatibility of any
two traditions of faith somewhat more in detail in my 'A Comparison of
Islam and Christianity as Frameworks for Religious Life'; but there I did
not develop adequately what I feel must be the basis for mutual comprehen-
sion among religious traditions: growth within tension, through persistent
dialogue. (That article was published in Diogenes, i960, but in so mangled
a form that it cannot be suggested for reading. The reprint by the University
of Chicago Committee on Southern Asian Studies, reprint series No. 10,
contains a fuller text and a list of essential corrigenda to the printed portion
of the text.)

On defining civilizations

In civilization studies — the study of the great cultural heritages (especially
those dating from the pre-Modern citied ages) — what may be called a
'civilization' forms a primary unit of reference. Yet the specification of
such units is only partly given by the data itself. In part, it is a function of
the inquirer's purposes.

Once society has become fairly complex, every people, even each sector
of the population within what can be called a people, has had a degree of
cultural self-sufficiency. At the same time, even the largest identifiable
group of peoples has never been totally self-sufficient. Even the cultural
patterns so large a group have in common will show interrelations with those
of yet more distant peoples. Social groupings have intergraded or overlapped
almost indefinitely throughout the Eastern Hemisphere since long before
Islamic times. If we arrange societies merely according to their stock of

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 31

cultural notions, institutions, and techniques, then a great many dividing
lines among pre-Modern civilized societies make some sense, and no dividing
line within the Eastern Hemisphere makes final sense. It has been effectively
argued, on the basis of the cultural techniques and resources to be found
there, that all the lands from Gaul to Iran, from at least ancient
classical times onward, have formed but a single cultural world. But the
same sort of arguments would lead us on to perceive a still wider Indo-
Mediterranean unity, or even (in lesser degree) the unity of the whole Afro-
Eurasian citied zone. In these circumstances, any attempt to characterize a
less extensive 'civilization' requires adopting an explicit basis on which to
set off one body of peoples from another as a civilization; but too often
such groupings have been taken as given, on extraneous grounds, and
characterizations have then been attempted without regard to the basis on
which the grouping was made.

We have yet to develop an adequate analysis of cultural forms for studying
the pre-Modern citied societies. Anthropologists have acquired some sophis-
tication in dealing with non-citied societies, and some of them have extra-
polated their methods into citied societies. Sociologists have learned to study
Modern Technical society, and generally societies of the Technical Age in
the light of it. But far too few since Max Weber have systematically explored
the periods and areas between — that is, from Sumer to the French Revolution.
This is partly because the lack of a tenable framework of world history,
which would supply an elementary sense of the proportions and interrelations
of the field, has hampered any comparative studies there. Anything may be
compared with anything else, but fruitful comparisons require relevantly
comparable units of comparison, which can be assured only through a sound
sense of overall context. In consequence, the questions posed about the
pre-Modern civilizations, and in particular about the Islamicate, have often
been irrelevant or misleading, and the answers they yield have been beside
the point or positively false.

It may be noted here that this lack of a proper world-historical framework
has probably arisen at least in part for want of a proper framework for
scholarly co-operation. What are commonly called 'Oriental studies' form
the larger part of what are better called 'civilization studies', including the
European heritage along with the others, since fundamentally the same
methods are involved in all cases, and the historical problems are all inter-
related. It is absurd for scholars in Islamics studies to be sharing conferences
with those in Chinese studies more readily than with those in Medieval
European studies.

It has largely been philologians who have — by default — determined our
category of 'civilizations' : a civilization is what is carried in the literature
of a single language, or of a single group of culturally related languages. This
notion has been presupposed by Carl Becker, Gustave von Grunebaum, and
Jorg Kraemer, for instance. It is not, in fact, a bad notion, to the extent

32 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

that my definition of a civilization in terms of lettered traditions is sound.
But it is not the same as what I am suggesting; it needs to be refined. In
crude form it has led, for instance (as we shall be noting), to an approach in
which everything carried in Arabic, including pre-Islamic pagan Bedouin
customs, is regarded as native and ancestral to the civilization that later
expressed itself largely in Arabic; while materials in Syriac, for instance,
produced in the mainstream of cultural development under the earlier
Muslim rulers and leading directly to central features in the urban life of
the civilization, are regarded as 'foreign' to it, and as 'influencing' it when
their ideas were 'borrowed' into it at the point when their exponents began
to use Arabic. The resulting picture of cultural development is, I believe,
erroneous. In principle, a field of study such as 'whatever culture happens
to be attested in Arabic documents' can be legitimate; but its relevance is
limited. If, for instance, we deal not with Islam but with Arabic as our
point of departure, so regarding Iranians as outsiders, we think of Bedouin
notions as 'surviving' while Iranian ones 'influence' the later culture from
outside. The Arabic culture of the High Caliphate then takes on two traits:
(a) suddenness; (b) a derivative character, as largely 'borrowed'. What a
difference in tone, if rather we should look at the problems posed by an
overlay of Arabic 'borrowing' upon Iranian and Syriac 'survivals'! Accord-
ingly, we must respect the challenge presented by men like Toynbee, who
defines his civilizations according to criteria based on inner cultural develop-
ment. When he divides what has been called 'Islamic' civilization among
three different civilizations, I believe he is in error, but he reminds us that
if we make it a single civilization we must give some reason why.

The reason for distinguishing a 'civilization' cannot be a single, universal
one, however; it must almost be special to each case. For no more than lan-
guage does any other one criterion necessarily determine a grouping that will
be worth studying as a major large-scale culture. Even a localized culture, at
least on the level of citied and lettered life, cannot be defined simply in terms
either of component traits or of participant families. In cross-section, a
culture appears as a pattern of lifeways received among mutually recognized
family groups. Over time, it may be more fully defined as a relatively auto-
nomous complex of interdependent cumulative traditions, in which an un-
predictable range of family groups may take part. It forms an overall setting
within which each particular tradition develops. But even within one
relatively local culture, some traditions — a given school of painting, say, or a
particular cult — may come to an end, and new ones may take their places. It
is not possible to distinguish, in any absolute sense, authentic or viable from
unauthentic or unviable traits in a culture, or even authentic from unauthen-
tic traditions. Yet a culture does have a certain integrality. The consequences
and the meaning of any given trait, inherited or newly introduced, will
depend at any given time on what implications it has for the ongoing inter-
action, the dialogue or dialogues into which it fits (or which it confuses) . The

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 33

consequences and ultimate meaning of any given particular tradition will in
turn depend on its implications for the cultural setting as a whole. These
implications will be more decisive, the more they touch the most persistent
and widely ramified features of the culture. Over time, then, what sets off a
culture as an integral unity in some degree is whatever makes for cultural
continuity in that particular culture.

On the wider and more rarefied level of what may be called a 'civilization',
cultural identity is even more problematic and what will make for continuity
is even less predictably formulable. We may indeed describe the most likely
situations in general terms which may seem to settle the matter. If we may
call a 'civilization' any wider grouping of cultures in so far as they share
consciously in interdependent cumulative traditions (presumably on the
level of 'high culture' — of the relatively widely shared cultural forms at the
urban, literate level of complexity and sophistication), then the shared
traditions will be likely to centre in some range of 'high' cultural experience
to which the cultures are committed in common. This may be a matter of
literary and philosophical as well as political and legal values carried in let-
tered tradition, with or without explicit allegiance to a given religious com-
munity. (Usually, lettered tradition is indissociable from the continuity of
written language; yet there need not be cultural identity except marginally
between two groups, especially in different periods, using the same language.
Many would refuse to put ancient Attica and Christian Byzantium in the
same civilization just because both used Greek and even read Homer. What
matters is the dominant lettered traditions, with their attendant commit-
ments, in whatever language.) When such major lettered traditions, then, are
carried in common, often there will be a continuity likewise in social and
economic institutions generally. All cultural traditions tend to be closely
interdependent. Often the integration within one area has been so marked
and the contrasts between that area and others so strong that at any given
time a demarcation line has been quite clear, and that line has tended to per-
petuate itself. Thus we get, especially in cross-section, the impression of
clearly marked civilizations parcelling out among themselves the Eastern
Hemisphere.

But this apparent clarity should not persuade the historian to take his
categories for granted. There will always be 'borderline' and 'anomalous'
cases which are quite as normal as the major groupings. It would be hard to
place such peoples as Georgians and Armenians unequivocally within any one
major 'civilization'. In any case, it cannot be clear in advance what sorts of
life patterns will in fact be found to be shared among the peoples forming what
can be called a 'civilization'. Each civilization defines its own scope, just as
does each religion. There may even be several sorts of basic continuity which
may overlap in range. Thus, depending on one's viewpoint, Byzantine
culture may be seen as continuing the ancient Hellenic tradition, or as part of
a Christendom briefer in time but wider in area ; and in each case there is a

34 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

genuine and effective continuity on the level of 'high culture' and its com-
mitments. Hence over a time span it often becomes a matter of choice —
depending on what sort of lettered traditions one specially wants to inquire
into — which among several possible delimitations will prove most suitable.
Then the scholarly treatment of the 'civilization' must differ with the
grounds for singling it out.

On determinacy in traditions

However a civilization be defined, it must not be hypostatized, as if it had a
life independent of its human carriers. The inherited cultural expectations at
any given time form part of the realities that members of a given society must
reckon with. They even put limits on what the most alert of those members
can see in their environment. But they have no effect except as they interact
with the actual environment and the immediate interests of all concerned.
The determinacy of tradition is limited, in the long run, by the requirement
that it be continuingly relevant in current circumstances.

Continuing relevancy is crucial to recall especially when cross-cultural
comparisons are being made. For instance, in an attempt to understand why
it was in the Occident that, eventually, technicalized society arose, scholars
have looked to the state of the Occident in the centuries preceding the trans-
formations. This can be done along two lines: by studying the special cir-
cumstances of the time when the transformations began and the special
opportunities open to Occidentals at that time; or by studying inherent
differences between Occidental culture and other cultures. In the latter case,
a comparative study of the High Medieval Occident with its contemporaries
is fundamental.

It is this latter case that has seemed the easier in the past. An adequate
framework of overall world history was lacking as a basis for studying the
special characteristics of the time of the transformation itself, whereas the
chief other societies were just well enough known as isolated entities to allow
specious global generalizations to be made about their cultural traits, traits
which could be contrasted to subtle traits traceable in the more intimately
understood Occident. Moreover, studying inherent traits in the Occident did
have undeniable relevance to a related question, often confused with the
question why it was the Occident that launched Modernity. The special form
that modern technicalization took, coming where it did, certainly owes much
to special traits of the Occident in which it arose. Since without adequate
world-historical inquiries it is hard to sort out what has been essential and
what accidental in technical Modernity, studies of what was special to
Occidental culture as such, which were assured at least some success in
accounting for the shape of Modernity as it actually arose, were mistakenly
supposed to have succeeded in accounting for where and when it arose.
Accordingly, scholars have been tempted to invoke, in accounting for the

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 35

advent of Modernity, the determinant effect of a fortunate traditional at-
titude or combination of attitudes in the pre-Modem Occident. Comple-
mentarily to that, often enough, they have invoked the 'dead hand of
tradition' to explain the 'failure' of other societies such as the Islamicate,
which are then compared, to their disadvantage, with the pre-Modem
Occident. The circumstances of the time when Modernity was launched have
been relatively neglected.

All attempts that I have yet seen to invoke pre-Modem seminal traits in
the Occident can be shown to fail under close historical analysis, once other
societies begin to be known as intimately as the Occident. This applies also to
the great master, Max Weber, who tried to show that the Occident inherited
a unique combination of rationality and activism. As can be seen here and
there in this work, most of the traits, rational or activist, by which he sought
to set off the Occident either are found in strength elsewhere also ; or else, so
far as they are unique (and all cultural traits are unique to a degree), they
do not bear the weight of being denominated as so uniquely 'rational' as he
would make them. This applies to both Occidental law and Occidental
theology, for instance, where he partly mistook certain sorts of formalism
for rationality, and partly simply did not know the extent among Muslims,
for instance, of a probing rational drive. But when the several traits prove
not to be so exceptional, the special combination of them that he invoked
as decisive loses its cogency.

It must also be noted that his method, as such, sometimes did not push
quite far enough. He sometimes depicted the attitudes he found as if they
were standing facts with automatic consequences, rather then processes that
never remain quite the same and have regularly to be renewed. Accordingly,
he could neglect the historical question of what it was that kept the attitude
in being once it had arisen ; and so he failed to see the full range of its inter-
action with other things, including with its own consequences.

The question of the relation of pre-Modern Occidental culture to Modernity
is a specially intriguing case of a much wider problem : the relative role in
historical development of traditional culture and of the current play of
interests. When it becomes clear that long-range historical change cannot be
adequately interpreted in terms of the initiative of great men or of direct
geographical or racial causation ; and when interpretation through the evident
moral level of the leading classes or even through immediate economic in-
terests proves to require explanation in turn of why the moral level or the
economic interests were as they were; then recourse can be had to explanation
by unevident but seminal culture traits. These seminal traits are supposed to
have latent implications, not visible in the earlier course of the society, the
consequences of which unfold at a later stage of the society's development —
if it may be assumed that the society has a determinate course of develop-
ment. Of the several sorts of seminal traits invoked, the most commonly
appealed to are inherited attitudes of mind, evaluations of what is good and

36 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

what bad. Thus in contrast to an Occidental inclination to rationalize and to
reinvest is posited an eternal Chinese inclination to tao-ize and to become
gentry; whereupon the failure of the Chinese to carry through an industrial
revolution is ascribed to their successful families' not persisting in industry,
but turning to other, more honoured, careers. (If the Chinese had been the
first to fully industrialize, they might have accounted for this also by their
wealthy families' tendency to become gentry — and so to sell their industries
to ever new blood, willing to innovate.)

I am sure that seminal traits may exist, though it is hard to pin them
down. But any evaluation of their historical effects must take into account
the full ecological setting of a given generation — that is, all the conditions
(including both geographically and socially given resources as well as current
interrelations with other groups) that would determine the effective advan-
tage of various possible lines of action and hence of attitudes that might be
adopted. Ideally, one should determine the points at which, under the given
conditions, additional investment of money, time, intellectual effort, etc.,
would yield diminishing returns. Such calculations would have to take into
account natural, man-made, and demographic resources, technical and
scientific alternatives available, and social institutions as given to that gen-
eration, including patterns of expectation, and what at that time these ex-
pectations depended on (that is, what it was that, at that time, might have
altered them). Such a listing would have to include the consequences of
ancestors' attitudes; but under the circumstances facing any given genera-
tion, the consequences of those attitudes need not come to the same thing as
the attitudes themselves. Even the outcome, in a given setting, of child-
raising techniques — the area where an unconscious past seems likely to
weigh heaviest — can vary strongly.

Attitudes like 'individualism', 'sense of personal vocation', or 'world-
negation' are hard to define closely enough for such purposes. It is easier to
trace the particular tokens of such attitudes ; and these can come to take on
quite opposite implications in a new setting. Thus the expectation in the
USA, that each nuclear family should have its own lawn-surrounded house,
which originally was doubtless a bulwark of certain aspects of individual in-
dependence, can lead, in certain sorts of 'organization-man' suburbia, to
bolstering social involvement and conformity. Or the exclusivity of the
Qur'an, with its rejection of the reliability of Jewish and Christian religious
witness, could contribute (by way of the self-containedness of the Qur'anic
exegesis) to the special universalism and tolerance of divergent traditions
(and not only of those of the People of the Book), which characterize some
strains of Sufism.

Indeed, whatever the situation may be in non-lettered societies, in every
complex society most relevant attitudes are to be found either among the
multiplicity of variant and practical traditions, or within that lettered tradi-
tion that has maximum prestige. Most temperaments and most possible

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 37

facets of experience that are to be found in any major tradition can be found
in corresponding traditions elsewhere. Accordingly, tradition can account for
almost anything. Thus for a time it was sought to prove that basic familistic
attitudes would prevent the Chinese from turning Communist; now the
Chinese bureaucratic heritage is shown to have made the Chinese peculiarly
susceptible to Communism.

Accordingly, it is wise to posit as a basic principle, and any deviation
from which must bear the burden of proof, that every generation makes its own
decisions. (This is perhaps a partial application of Ranke's principle that all
generations are morally equidistant from God.) A generation is not bound by
the attitudes of its ancestors, as such, though it must reckon with their conse-
quences and may indeed find itself severely limited by those consequences in
the range of choices among which it can decide.

The difference between major traditions lies not so much in the particular
elements present within them, but in the relative weighting of them and the
structuring of their interplay within the total context. If this structuring
remains relatively constant (in the very nature of tradition, it cannot remain
absolutely so), it will be because the predisposing conditions remain relatively
constant, and because they are further reinforced by the institutionalizing of
attitudes appropriate to them. Such institutionalizing can indeed be crucial
in making the predisposing conditions fully effective: e.g., the mercantile bias
of the Irano-Semitic cultural traditions, already so visible in the develop-
ment of the monotheisms, was given fully free play only under the auspices
of Islam. The triumph of Islam was made possible by its special adaptation
to that bias, but its triumph in turn allowed that bias to determine the
subsequent course of Irano-Semitic history. Nevertheless, the consequences
of such institutionalization cannot reach very far in independence of the pre-
disposing conditions. They can allow a tendency already the strongest in the
field to become fully effective, and they can then reduce fluctuations that
might result from variations in the underlying conditions, so that a temporary
or a local deviation from the general norm will not produce a total cultural
disruption. But if altered basic conditions long persist, the corresponding
attitudes and their institutionalization will soon be changed to match.

Historical change is continuous and all traditions are open and in motion,
by the very necessity of the fact that they are always in internal imbalance.
Minds are always probing the edges of what is currently possible. But even
apart from this, we are primarily human beings with our personal interests
to pursue, and only secondarily participants in this or that tradition. Any
tradition must be regularly reinforced by current conditions so that it
answers to current interests or it will perish by drying up — or be transformed
into something relevant. Whatever unity of patterning we may be able to
discover, as to primacy of orientations or as to validation of norms of orga-
nization — whatever sense of common style we may find in the culture, that
is — may be very pervasive and persistent and yet be essentially fragile. As

38 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

soon as new positive possibilities open up, the unity of patterning is quickly
vulnerable. To the extent that a homogeneous and compelling style is at-
tained, in fact, it must be regarded as a delicate flower, not a tap root ; it is
not something imposed by cultural necessity, though the range of potentia-
lities may be given so, but something achieved by creative effort.

A special word has to be said about one of the crudest, yet remarkably
pervasive forms of hypostatizing a cultural tradition — or, in this case, a
whole series of them. The misimpression that 'the East' has latterly been
awaking from a 'millennial torpor' is still remarkably widespread. It results,
of course (like the term 'East' itself), from the profound ignorance of world
history not only among modern Westerners but among others as well, whose
eager vaunting of the antiquity of their institutions was taken at face value
by Westerners.

We may single out two types of scholar who have reinforced the mis-
impression. Western tourists, whose moods played a large role even in
scholarship, easily mistook the exotic for the immemorial, and were necessar-
ily blind to subtler institutional changes. Their impressions, then, were
dignified into learned theses, sometimes of a racialist hue, by scholars be-
mused by the spectacular progressiveness of their own West, and ready to
write off other societies as irrelevant. Reading back the recent Western pace
of activity into the earlier Occidental past by a foreshortening of time-spans
in the distance, and unaware that in other parts of the world there was a
comparably active past, Western scholars assumed that the comparatively
slow pace of technical and intellectual development which they could per-
ceive in the nineteenth-century world abroad amounted to no development
at all, and marked a difference of race and place rather than one of age.

But other Western scholars — well represented in 'area studies'- — have
confirmed the misimpression by an opposite error. While more or less
recognizing the comparability of pre-Modern Western and non-Western
societies as to degree of cultural activity, they have blanketed all pre-
Modern areas under the common term 'traditional', the misleading tendency
of which we have already seen, as if all had been asleep together (save in
certain periods of undeniable florescence) — rather than all awake together.
As we have noted, the degree to which pre-technicalized and even pre-literate
peoples have been bound by the 'dead hand of tradition' has been greatly
exaggerated. Among Muslims, at any rate, the major institutions of each age
can be shown to have their own functional justification in their own time:
Muslim social decisions, even under the conservative spirit, were made not
primarily out of deference to the past but as meeting concrete practical
interests of dominant social groups. Whether it is the 'East' or the 'pre-
Modern' that is being misperceived, the postulate of essential changelessness
obscures the important question of how the particular posture in which
various peoples happened to be at the moment of the Transmutation affected
their destiny under its impact. There is too ready an answer to the question

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 39

of why 'reform' efforts so often failed : the 'tradition-bound' lands were ruled
by blind conservatives. Some are thus spared the trouble of discovering what
very practical and alert statesmen those 'tradition-bound' men often were . . .
except in the case of the Japanese, who are gratuitously labelled 'good
imitators'.

On the history of Islamics studies

Only gradually have scholars come to recognize the relevance of the scholar's
perspective in delimiting his field, enough to make the perspective conscious
and so keep it relatively under control. Historically, scholars' notions of
'Islam' as a field have been rather arbitrarily determined by a series of
political and other extraneous circumstances. These notions still have their
consequences. Almost all stages of the historical development of modern
Islamics 18 studies are represented in the works that an inquirer must still
consult in the library. What is more, many of these stages are still repre-
sented in studies made in the mid-twentieth century. Hence even the
relatively casual student of things Islamic should be aware of the history
of Islamics studies. This will allow him better to appreciate the relevance
of the individual scholarly studies to whatever his own interests are, as
well as put him on guard against various endemic but avoidable errors that
have come to prevail in the field.

Because of the cultural circumstances of the Modern Technical Age,
Western scholarship has been the chief channel for studies of a Modern
type in the Islamics field at least until very recently. Western scholarship
entered the Islamics field above all by three paths. First, there were those
who studied the Ottoman empire, which played so major a role in modern
Europe. They came to it usually in the first instance from the viewpoint
of European diplomatic history. Such scholars tended to see the whole of
Islamdom from the political perspective of Istanbul, the Ottoman capital.
Second, there were those, normally British, who entered Islamics studies
in India so as to master Persian as good civil servants, or at least they
were inspired by Indian interests. For them, the imperial transition of
Delhi tended to be the culmination of Islamicate history. Third, there were
the Semitists, often interested primarily in Hebrew studies, who were lured
into Arabic. For them, headquarters tended to be Cairo, the most vital
of Arabic-using cities in the nineteenth century, though some turned to
Syria or the Maghrib. They were commonly philologians rather than
historians, and they learned to see Islamicate culture through the eyes of
the late Egyptian and Syrian Sunni writers most in vogue in Cairo. Other
paths — that of the Spaniards and some Frenchmen who focused on the
Muslims in Medieval Spain, that of the Russians who focused on the

18 On the use of the term 'Islamics', see the section on usage in Islamics studies
below.

40 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

northern Muslims — were generally less important. All paths were at one in
paying relatively little attention to the central areas of the Fertile Crescent
and Iran, with their tendency towards Shi'ism; areas that tended to be
most remote from Western penetration. Unless perhaps in Russia, studies
of the central areas still tend to be neglected except for the earliest centuries ;
Islamdom is rarely seen from such a perspective.

For a long time, in any of the paths it took, Modern Western scholarship
was largely a matter of translating the results of pre-Modern Muslim scholar-
ship and adapting them into Occidental categories. Improvement in Western
scholarship was largely a matter of moving from later, secondary Muslim
sources to earlier, more nearly primary ones; new editions of more ancient
texts tended to be the most important scholarly events. At the same time,
the Western scholarly viewpoint changed in response to shifts in viewpoint
which took place during the nineteenth century among Muslims themselves :
shifts which emphasized the importance of the early Arabic period and the
Shari'ah associated with it (as against the more recent Persianate and
Sufi tendencies, which were being rejected as decadent). Both these currents
of change tended to emphasize the earlier periods and the Arabic documents
as the object of the best scholarly work and the focus of scholarly inter-
pretations. In consequence (especially with the decline of Istanbul in inter-
national importance after 1918), the Cairene path to Islamic studies became
the Islamicist's path par excellence, while other paths to Islamics studies
came to be looked on as of more local relevance : via Istanbul one became
an Ottomanist, for instance, but no longer a scholar of Islamics as such.

All this, then, reinforced the Arabistic and philologistic prejudices which
resulted anyway from several European tendencies : the interest in a Semitic
'race' (set over against an Indo-European 'race' represented by the West)
which was expected to illuminate the Semitic Biblical background; the
interest in 'origins' of supposedly isolable cultural entities, with which
nineteenth-century Western scholarship was obsessed; the concern with
Mediterranean (and hence largely Arab) Islamdom, as nearest to Europe
and most involved with its history; and the philological tendency to learn
Arabic as the most essential linguistic resource and to stop there rather
than going on seriously to other languages. This Arabistic and philological
bias is reflected in book after book and article after article; not least in
the Encyclopaedia of Islam, where many entries discuss more the word (usually
in its Arabic form even if it is derived, say, from Persian) than the substance;
and present data for Egypt and Syria as if for Islamdom as a whole. (For
a consideration of the problem of getting past this situation, see the
section on usage below.)

Meanwhile, however, other changes had been modifying Western historical
scholarship generally, and with it Islamics studies. It is especially in the
twentieth century that Islamicists have been going beyond the results of
older Muslim scholarship to pose their own questions and derive their own

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 41

answers from the documents, now pulled apart as bits of evidence rather
than copied as authorities. More slowly, scholars are learning even to get
beyond categories derived from the Occident, partly by learning to use with
more precision categories used by Islamicate writers themselves, and partly
by learning to use relatively neutral categories disciplined by wider studies
of world history and of human society and culture generally. This task is
as yet far from complete, however, even in the work of the best scholars.

Apart from this, present-day Islamics studies still suffer from the
philologism of their past. Their Arabistic bias, with the neglect of the more
central Islamicate areas, is only gradually being overcome. (Yet with the
dropping of the training in general Semitic studies that Arabists used to
have, a great advantage of the old philologism is being undermined: its
ability to integrate Arabic with older Semitic — especially Hebrew and
Aramaic — studies and sometimes even with older Iranian studies.) Perhaps
even more important, Islamics studies have tended to be concerned, above
all, with high culture, to the neglect of more local or lower-class social
conditions; and within the high culture, to be preoccupied with religious,
literary, and political themes, which are most accessible to a philological
approach. Hence it is important to point out such journals as Comparative
Studies in Society and History and Social and Economic History of the Orient,
which have stressed other sides, and in both of which Gustave von Grunebaum
has taken a leading role in encouraging a good contribution to Islamics
studies. Claude Cahen has been the most effective Occidental writer on
social and economic questions in many dispersed articles, to be traced down
through the Index Islamicus.

I will here illustrate the problems that arise from too great a reliance
on a philological outlook by way of offering some caveats on the work of
scholars on the period immediately before and after the rise of Islam, when
lack of broadly based data encourages an unhistorical use of what there is.

The special role of old Arab families has given rise (in combination with
certain more extraneous circumstances) to a tendency on the part of many
scholars to interpret the development of Islamicate civilization from an
Arab, even an Arabian point of view. Identifying 'Islamic culture' as 'culture
appearing in the Arabic language', they will treat all pre-Islamic Arabian
elements (i.e., those found in the Arabian peninsula) as native to Islamic
culture, and will think of an Arabian Bedouin folkway as 'lost' or 'dropped'
if it is not found among later Muslims in the Fertile Crescent. Correspond-
ingly, they will treat Syriac, Persian, or Greek cultural elements as 'foreign'
imports into Muslim Arab life, despite the fact that they formed the ancestral
traditions of most of the Muslim population, and even of most of the educated
and culturally privileged section of that population. The term 'pre-Islamic'
for some scholars means strictly 'pre-Islamic Arabian', not pre-Islamic
from Nile to Oxus generally wherever Islam was established.

When one focuses attention, as one often must, on the Arab Muslims

42 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

and especially on the creatively concerned minority among them, this
standpoint is largely relevant ; for from the standpoint of the ruling Arabs,
and even of their more ardent converts, everything non-Arabian was
'foreign'. What the observer might call the Arabs' gradual assimilation into
the established cultural patterns among which they came to live appeared
to the Arabs, rather, as a gradual assimilation of external cultural elements
into an on-going essentially Arab society. It is a legitimate question, how
such non-Arabian elements were assimilated among Arab families.

But if one is to understand the larger scene in terms of which even the
concerned men of old-Arabian family were working, this Arab-centred or
'Arabistic' viewpoint can be misleading. Unfortunately, this Arabistic
approach has, in fact, often been carried too far by philologists, for whom
the language group (in this case, Arabic) is the key unit of all historical
study, and for whom the origin of terms is occasionally more fascinating,
or at least more accessible, than the origin of the actual institutions to
which they are applied. In fact, this approach may be called the conven-
tional one ; it has often imposed its terminology even on those who might
otherwise be relatively free of it. Yet if not balanced with other points of
view, the Arabistic viewpoint can put developments into a false light,
arousing false expectations and raising false questions. The reader of studies
in the field must be ever on guard against being misled in this way; the
Arabistic bias in studies of early Islam has been remarkably hard for even
the best scholars to get past.

For instance, if comparisons of conditions before and conditions after the
advent of Islam are made between pre-Islamic Arabia on the one hand
and Syria or Iran in Islamic times on the other, they can be revealing of
what happened to those families that formed the Arabian element in the
new Islamic societies. But since those families had undergone not merely
Islam but also a major migration and a great rise in social status, any
comparison may not tell much about Islam itself or its culture, unless it is
balanced with comparisons between pre-Islamic Syria, say, and Syria
under Islam; and between pre-Islamic Arabia and Arabia itself in Islamic
times. Otherwise, differences may appear to result from Islam which are
matters of politics and of geography. Yet such balancing has rarely occurred
in published work.

To such a philological bias is often added the old unexamined assumption
of identity among the three ethnic criteria: patrilinear race, language, and
cultural heritage. 'Arabs' by language and 'Muslims' by heritage are often
identified, even at much later historical periods, as almost a single category,
in which Arabian 'race' is thought of as normal, even though it is ac-
knowledged that the 'exceptions' have been far more numerous than the
'normal' cases, such 'exceptions' being duly noted as such. At this point,
historical inquiry can be seriously thrown off. Writers are to be found,
for instance, asking how Greek elements entered (as non-Arabian) from

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 43

'outside' not merely into a few Arab families but into the whole Arabic-
using society from Nile to Oxus, in which (in fact) they were already present
by inheritance; while the same writers fail to inquire how such elements
were avoided for so long in the education of upper-class Muslims. This latter
question in turn leads to the wider and more serious question (which too
purely philological a scholar will scarcely know how to set about asking),
how the Arabic language and with it so much of the Arabian background
managed to emerge as a cultural framework in a society where they were
so greatly disadvantaged. How did Arabian foreign elements — alien at
several points even to Islam itself as conceived by Muhammad — come to
receive such relatively ready acceptance among the Semitic and Iranian
peoples?

The answer partly lies in the development of those peoples themselves.
The Islamicate civilization may be seen as the latest phase of the Irano-
Semitic culture which goes back, in the lands from Nile to Oxus, to
Sumerian times. The very vision of Islam grew out of that heritage; even
the mercantile-nomad understanding which led to the Arab conquest
was not entirely alien to it. Islam as an Arab creation was able to come
to flower in those wider lands because it answered to circumstances that
were already determinative there. The Irano-Semitic cultural traditions
showed a long-term tendency, within what remained overall an agrarian-
based cultural context, to shift from a more agrarian base for high culture
toward a more mercantile base for it. What was distinctive in Islamicate
civilization grew largely from the special role in it of traditions linked to
the mercantile classes. It brought to culmination what had long been
developing.

The antidote to the Arabistic bias ought to lie in marshalling the data
on the rest of the society other than the few Arabian families. But unfor-
tunately, this has not proved easy. The few ambitious attempts to do it,
moreover, have themselves suffered from a philological approach in a
different form.

Sasanian religious and social history must be reconstructed from archaeo-
logical and indirect textual sources, with very few major literary witnesses
from the period itself. Aramaic, Greek, or Armenian texts view affairs
largely from the outside and marginally. The Pahlavi texts are often suspect
as having been edited, at least, in Islamic times, or they must even be
reconstructed from Arabic and Persian translations and adaptations. Even
when we do have undoubted Pahlavi texts from Sasanian times, the original
script was so tricky and the manuscript tradition has been so defective that
reading the texts must be left to philological specialists. In consequence,
few scholars have entered the field and those who have been tempted to
indulge in rather wild philological speculation, building much on shreds of
verbal detail. Since Arthur Christensen's work, for instance, two of the
most spectacular writers have been Robert Zaehner (notably, Zurvan, a

44 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

Zoroastrian Dilemma [Oxford, 1955]) , and Franz Altheim, who summarizes
in Utopie und Wirtschaft: eine geschichtliche Betrachtung (Frankfurt/Main,
1957) the more questionable results of his more detailed work {with Ruth
Stiehl), Ein asiatischer Staat: Feudalismus unter den Sasaniden und ihren
Nachbarn (Wiesbaden, 1954). Both illustrate the pitfalls of philological
limitations even in their excellences.

Zaehner's work is very informative in detail and suggestive in some
correlations, yet he needs to point out the difference between a verbal
formulation and its existential, experiential meaning in real life between
the doctrine of a religion and the mood of a poem. Consequently, his descrip-
tion of the evolution of Zervanism, which seems to have been a Mazdean
school of thought which did not survive, is unconvincing in detail, for one
can usually think of less improbable explanations of the particular points.
I thus find it unconvincing as a whole. The like tends to be true of his work
in Islamics studies, where he takes a few schematic notions and builds a
whole typology of faith on the presence or absence of them in the verbal
formulations of a given thinker.

Altheim's work is likewise informative and often suggestive; but here
again far too much is built on single reconstructions of textual passages,
while human probability is flaunted. Nushirvan's tax measures on the
surface, as Altheim points out, generalized taxation in money rather than
in kind and therefore suggest an increasing strength in the monetary and
mercantile aspects of the economy. But by a series of ingenious and most
improbable correlations, Altheim finds, even in this, evidence for the
reverse: Nushirvan's taxation was introducing control of the economy
rather than mercantile market freedom; his lesser dependence on the
traditional landed gentry likewise meant closer controls of the society, and —
by creating a new lesser gentry — put the central government in more
immediate dependence on the land. The whole, Altheim maintains (by way
of weak evidence on castle building and anachronistic misconstruction of
Muhammad's work), led to strengthening of the non-monetary 'natural'
economy and so (by interaction with Byzantium) to the Middle Ages. At
almost every point, a more adequate awareness of post-Axial agrarianate-
level social conditions as such would have suggested better alternative
interpretations.

Both Zaehner's and Altheim's systems make such sense as they do by
incorporating certain uncriticized preconceptions about world history which
allow them to overlook alternative possibilities in their philological recon-
structions. Zaehner speaks, without a word of apology, of a Zoroastrian
'church', of 'orthodoxy', of 'sects' of which he takes for granted his
Zervanism must be one, though he gives no grounds for supposing that
such phenomena, in the sense he presupposes, were in fact present in that
period. Altheim, correspondingly, takes for granted the underlying stereo-
types of 'Orient' and of 'Middle Ages', without which his scheme would

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 45

have no plausibility. His 'Orient' or 'East' is an eternal entity, from ancient
times down to the present Soviet Union, always basically backward and
irrational but capable of learning from the essentially rational and progres-
sive 'West' and of learning so well as to force or drag the latter into a Dark
Age. For him, Mazdak was an ancient Marx, the forerunner of Islam as
was Marx of the Soviet system. This parallelism is again made easier by a
stereotype: it has become conventional to refer to Mazdak and other
reformers as 'communistic', advocating 'community' of property and of
women. This terminology Altheim continues to use although he himself
makes it clear enough that neither community of property nor community
of women was involved, but rather their redistribution, in certain cases, to
other private individuals. He seems never to have thought through these
various stereotypes to consider what their meaning could really be. Such
confusion as to basic historical categories plagues Islamics studies all
through, but is nowhere so clear cut (or so disastrous) as in the Sasanian
period. The general scholar is forced, even more than in the Islamic periods,
to reconstruct by educated guesswork for want of adequately grounded
scholarly guidance.

USAGE: REVISIONISM IN SCHOLARLY TERMINOLOGY

If one must consciously choose and face the implications of one's approach
to a civilization, so must one also choose and face the implications of one's
terms, selecting them relevant to the questions one is asking. In using such
words as 'Oikoumene' and 'Axial Age', 'Islamdom' and Tslamicate', I have
preferred to introduce new usages or adopt relatively rare ones rather than
use terminology now conventional. I have done this with reluctance,
recognizing that the historian, like the philosopher, has a special obligation
to be intelligible to the layman because of the human immediacy of the
questions he is dealing with. Nevertheless, some special terms and usages
are necessary. In many disciplines, scholars would not dream of taking
their terminology from the street. Even if they do not fully succeed in
agreeing upon a given set of terms, they recognize that it is essential for
each writer to use his terms with precision, and that an attempt to accom-
modate oneself to popular usage as reflected in a dictionary must be
disastrous. Too often, historians (especially in the field of Islamics) still try
to avoid recognizing such a necessity and are satisfied to be guided by
whatever is 'common practice'. They note that often terminological dis-
cussion can descend into pettifogging, and that the nature of their field
prevents historians from building up a single total body of terminology in
which all cases are provided for. They hope, therefore, that terminology
will take care of itself; but it does not. The responsibility remains for
selecting minimally misleading terms and for defining them precisely.

Terms are the units by which one constructs one's propositions. The

46 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

terms one uses determine the categories by which one orders a field — or
at least all those categories that are not the immediate focus of one's
inquiry. The categories one presupposes, then, necessarily delimit the ques-
tions one can ask — at least all the constants implied in the questions,
apart (again) from the special point of focus. The questions posed, in turn,
determine what answers will ultimately be reached when the questions, as
posed, are pursued. The story of scholarly error is largely one of questions
wrongly put because their presuppositions were wrong; correspondingly, the
story of scholarly achievement can almost be summed up in successive
refinements of terminology.

There are two approaches to conventional misusages : the admonitionist
approach and the revisionist. The admonitionists, admitting a given usage
or practice is misleading, prefer to maintain the continuity of communication
which even false conventions make possible, but to add a warning that such
and such a usage or categorization must not be taken in the most likely
way. The revisionists prefer to replace outright the conventional misusage
(or biased categorization) with a sounder one.

Those who have not had time to verify that the conventional error is
as unmitigated as the revisionists claim, or that the offered replacements
are not themselves tainted with unanticipated further error, almost neces-
sarily incline to wait and meanwhile have recourse to a caveat. But even
some who grant that the revisionist point is quite sound may feel that the
advantages of change do not outweigh the disadvantages which come with
any break with continuity.

As will appear especially from my defence of the term Tslamicate', in
the section below on usage in Islamics studies, in this work I have been
insistently and almost without exception revisionist where I saw occasion
arise. Whatever concessions need to be made, I feel that the categories and
terms arising from the Arabistic bias in particular need drastic revision.
But there are other cases where conventional misusages tend to reinforce
natural misconceptions. For instance, to put 'land assignment' (or 'land
grant' in some situations, perhaps) in place of the usual term 'fief in the
conventional discussions of what is admittedly miscalled 'Islamic feudalism'
will usually have one of two effects. Where the discussion happens to be
sound, the change throws the points being made into sharper relief — and
often makes many of the clarifications prove superfluous. But remarkably
often it shows up inadequacies and even absurdities in the conventional
discussion and its presuppositions. In the latter case, to hide from the
consequences of a more accurate usage, on the grounds of convention, does
not serve good scholarship.

The most common conventional errors tend to be bias-reinforcing errors,
which is, perhaps, just why they are clung to. With such errors, human
nature is such that a caveat does very little good. It is more than the divided
human attention can do to keep in mind a caveat that runs against one's

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 47

favourite presuppositions if those presuppositions are constantly reinforced
by the very terms one uses. In such a case, new terms and new practices
alone can take effect; the old, even amended, cannot usually transcend
themselves. In any case, in Islamics studies and in civilization studies
generally, the inadequacy of our studies up to this point is so great that an
attempt to maintain continuity is doomed to failure: we are still almost in
the 'pre-historic' period of scholarship. Continuity with old first approxima-
tions is bound to be of minimal value.

In fact, historians have already used some care in glaring cases. In the
field of Islamics we have got rid almost altogether of 'Moorish' and 'Turanian'
and 'Saracen' among specialists because each of these terms had come to
carry, as a category, implications too hopelessly confused for reform to
seem worthwhile. But far more debris remains to be cleared away. For
instance, in the case of terms for areas, European historians would not
think of analyzing past conditions in terms of current political boundaries,
positing a historic Switzerland, Austria, Germany, Belgium, and France
in their mid-twentieth-century limits; yet many Islamicists try to discuss
past periods in terms of 'Persia', 'Afghanistan', 'Syria and Palestine', and
even 'Lebanon', all of which refer to strictly modern entities, at least in
the meaning usually assigned them. We must face the fact that such
terminology will not do. It will not do in discussing particular historical
events where, fortunately, it is often possible to refer to territorial states
of the time. But it will also not do when describing long-term developments
and trends in given areas, for current boundaries rarely set off areas with
any such inherent unity as will allow for useful discussion. We must recon-
cile ourselves to using area terms based not on the political situation of
any period, but on more enduring criteria relevant to the discussions at
hand.

To take one unusually illustrative example, the coastlands between the
Mediterranean and the Syrian desert shared much in common and often
need to be referred to as a body. If we do not use the term 'Syria' for that
area — including what is currently Lebanon, and Palestine in its old, largest
definition, but excluding the eastern parts of what (by French fiat) is now
the Syrian republic — we must invent some other term. Most writers
recognize this fact, but — accepting current political usage of terms as pre-
emptive — try to express themselves somehow by means of them. First off,
though, the name 'Israel' cannot be used in any designation because of
current implied connotations. Even a periphrasis in 'Syria including Lebanon
and Palestine', however, is not only too clumsy but inaccurate, since it does
not exclude enough. The compound 'Syria and Palestine', tacitly including
current Lebanon in 'Syria' (since on the interwar maps they were both
coloured French) is a formation not purely political in inspiration. Partly
it reflects Western sentiment about the special place of Palestine, which
was to be set apart from the rest of Syria; but in net effect it amounts to

48 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

an interwar compromise which exaggerates the place of southern Syria in
the Islamicate scene without satisfying those who go purely by today's
political map. Hence some decide to degrade the old term 'Levant' to mean
only this smaller area; but then we will have to provide somehow for the
wider usage which 'Levant' used to serve; and in any case we will be
faced by French restriction of 'Levant' still further to cover just the two
states under French mandatory control. Such yielding to current political
and sentimental fashions in the end usually leads nowhere. Sometimes
political fiats and the language of the street must simply be defied and
terms be used on the basis of historical needs alone. In this case, unless
some other term can be found which the politicians will not proceed to steal,
it seems best to continue to use the term 'Syria' in the old sense it most often
had before 1918, which happens to be just the sense we need. We may indeed
concede 'geographical Syria' or 'the Syrian lands' or the like for the noun,
as avoiding a clash; but the adjective anyway must remain 'Syrian'. The
reader must be told to adjust himself! In other cases, special terms must
be invented. Sometimes I have had recourse to adding the article, where
an original Arabic article has made it appropriate, to indicate reference
to a geographic area rather than a modern state; e.g., 'the Iraq' for the
more southerly part of the current Iraqi state. But here, as with phrases
in 'lands' or 'area', there is no help for the adjective.

USAGE IN WORLD-HISTORICAL STUDIES

In the broad field of world history, the terminology of the street is especially
misleading, for it reflects consistently a strongly ethnocentric Western
view, radically distortive of the reality. The major terms for area and
period will suffice as illustrations. The periodization in 'Ancient', 'Medieval',
and 'Modern' has been attacked by innumerable historians as inadequate
for a fair long-run view even of European history; and while it is sufficiently
vague so that at least the second and third terms seem adjustable to any
area, yet overall it is still more distortive of the world scene than of the
European. Far worse is the geographical terminology. 'Europe', 'Asia', and
'Africa' referred initially to the north, east, and south hinterlands of the
Mediterranean. When the sub-Saharan region is lumped with the southern
littoral of the Mediterranean, and everything east as far as China and Japan
with the Levant, the resulting monstrosities would look like bad jokes were
they not made use of every day (even for statistical purposes) as if they
were real entities with real characteristics of their own. As it is, they are
vicious historical distortions. 'Europe', if taken loosely, is a more defensible
concept; yet attempts to trace an eastern 'boundary' for Europe through
the Aegean (the two sides of which have always had basically the same
cultural and historical features) and along the Urals (which have never
once served as either a political or a cultural boundary) would also be

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 49

easily dismissible were they not taken seriously and even inscribed on
popular maps.

The reason such terms persist, of course, is that on one level they do
serve a use. In the case of 'Europe' and 'Asia', at least, the artificial elevation
of the European peninsula to the status of a continent, equal in dignity
to the rest of Eurasia combined, serves to reinforce the natural notion,
shared by Europeans and their overseas descendants, that they have formed
at least half of the main theatre (Eurasia) of world history, and, of course,
the more significant half. Only on the basis of such categorization has it
been possible to maintain for so long among Westerners the illusion that the
'mainstream' of world history ran through Europe. (The acceptance of
such terms by non-Westerners too is a sign of their continuing cultural
dependence on the West.) The other major pair of popular world-historical
conceptions, 'West' and 'East' (or 'Orient') form a variant on the pair
'Europe' and 'Asia' and serve the same function of reinforcing Western
ethnocentric illusions.

It is, of course, precisely because of this strongly emotive unconscious
function that the careful pre-Modernist historian, at least, should never use
the concepts 'Asia' or 'Orient' but should refer precisely to the more limited
area he actually has in mind (the area is always in fact more limited, if
he is not just indulging in uncritical generalizations). But unfortunately
it is often just historians who have been misled most drastically into false
statements because they take those categories seriously. The key point
is to say what one means: 'Semitic', Tslamicate', Tndic', Tndic and related',
'Far Eastern', perhaps sometimes Tndic and Far Eastern' ; or else 'exotic',
'alien to the Occident', 'non-Western', perhaps 'Islamic and Monsoon
Asian', 'non-Western civilized'; or else 'indigenous', 'local', 'non-tech-
nicalized', or just 'other'! Even in the Modern period, when the non-
Western citied lands have had something in common precisely in not being
'Western', the terms 'Orient' and 'East' have connotations sufficiently
unfortunate in several ways to make them suspect, though I find the term
'West' useful in the Modern period for all those on the 'European' side of
the development gap.

My world-historical terminology is made clear, for the most part, as I
go along. Many of the terms or phrases I use in a distinctive sense are also
listed in the Index or the Glossary. Here I can explain a few choices.

The Occident is for me precisely western or Latin Europe and its overseas
settlements. If I mention the 'European region', it is with reference to the
extended northern hinterland of the Mediterranean, including the Anatolian
peninsula; if, especially in compounds, I use 'Europe' for short, in a pre-
Modern context, at least, I have in mind the same region. In a Modern
context I may use it for European Christendom instead. The term Tndic
region' likewise seems safer, when convenient, than just 'India' for the
areas southward of the Pamirs and Himalayas; and it is less pretentious

50 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

than the politically retroactive 'Indo-Pakistan subcontinent'. The (Eurasian)
'Far East' I use for the area of primary Chinese cultural influence — including
(for instance) Vietnam but not Cambodia. I have chosen the term
'Oikoumene', in a sense similar to that latterly used by Alfred Kroeber
not just as an area term but to refer to the Afro-Eurasian agrarian historical
complex as having a distinctive interregional articulation in an ever growing
area; there seems to be no other term for this complex at all. (The form
'Oikoumene' allows better than 'Ecumene' for an adjectival form distinct
from 'ecumenical'.) The term 'citied society' (i.e., containing cities, as distinct
from 'urban' which refers to the cities themselves) has the advantage over
'civilized' of avoiding, in certain contexts, any invidious implication of
degrees of refinement of manners; for most purposes it suffices without
further modifiers. The term 'agrarianate' (cf. note 3 in chapter 1 of Book
One) has the advantage over phrases like 'pre-modern' or 'traditional' of
being distinctly set off, as well from Modern technicalistic as from pre-
agricultural society. The single antithesis 'traditional-modern' not only
oversimplifies historically what is chosen as a contrast to 'Modern' but
underplays the dynamic nature of what is commonly thought of as tradition,
and it definitely ignores the active role of 'tradition' in the most Modern
society. The articulation of the long 'Agrarian Age' from Sumer to the
French Revolution as pre-Axial, Axial, and post-Axial suits most general
purposes for which terms like 'Medieval' would serve and gets round the
question whether to include the world-historical period from Columbus
to the French Revolution under the label 'Medieval', since from a world
viewpoint it is undeniably still post-Axial Agrarian, though within the
Occident it initiates Modernity. The important though shorter periods
required within the post-Axial, such as the age of the dominance of the
confessional faiths, can be referred to ad hoc.

Modern

Various terms are now used in referring to the distinctive complex of cultural
traits that have played a decisive role in human society since about the
generation of 1789. Most of such terms are appropriate in one context or
another. A first set of terms depends on the recentness of these traits and
on the fact that they do not remain constant, but must always be brought
further up to date. The age characterized by these traits (together with
that period which, within the Occident, can be regarded as leading up to
them) is usually called 'Modern' ; the traits can be summed up as 'Modernity',
and adoption of them, as 'Modernizing'. A second set of terms refers to the
high degree of economic exploitation of resources which is also a funda-
mental characteristic. A society lacking the traits in question is called
'undeveloped' or 'underdeveloped' and the acquisition of such traits is
called 'development', which properly should refer strictly to technical

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 51

development as applied to exploitation of resources, but can be generalized
to all the necessarily related traits. A third set of terms has a more precise
application. Because a key trait is technical rationality in the sense of
subjecting all behaviour to calculation according to presumedly objective
ends without interference from arbitrary tradition, the acquisition of the
traits generally can be called 'rationalization'. Finally, some refer to
acquisition of the traits in question as 'Westernization' because they were
first developed in western Europe, and because acquisition of them appears
to make any group seem like western Europeans.

In this study, usage on this point needs to be more effectively differentiated
than is customary. Use of the term 'modern' commonly presupposes use
of the term 'traditional' as a lumping term for all social forms not charac-
terized by the given complex of traits. This latter usage is very unfortunate
in as much as within the 'non-Modern' social forms there are often important
distinctions between what is and is not traditional, even in the sense of the
term — i.e., immemorially customary — intended by those who use it. For
instance, there has been a dynamic distinction between the Shari'ah of the
textualistic hadith-minded 'ulama', always fighting popular ways, and more
'traditional' custom which tended to continue popular ways; and such a
distinction may be of crucial import just in the present context. Again,
common current usage can result in calling 'traditional' certain political
or economic patterns that are clearly not characteristic of a 'Modern' form
of society in our special sense, but which may have developed only in the
nineteenth century and in response to the presence of 'Modern' conditions
in the environment. Such usage invites confusion on a very crucial aspect of
the history of our times by confounding these latter-day intrusions or
expedients with more genuinely 'traditional' traits. But even the term
'Modern', though useful in many contexts where it happens to be unam-
biguous, is not satisfactory as a precise technical term for the traits in
question, if only because it is so often necessary to distinguish between what
is up to date and what is out of date at any given point within the process
of Modern change. Though in this work I do sometimes use 'Modern'
(capitalized) to refer to what more properly I call 'technicalistic' or 'of the
Technical Age', yet it seems best as far as possible to retain 'modern' for
the up-to-date as such, using it in a more epochal sense ideally only when
it contains just that specific element of relativity and even of normative
implications — in other words, the element, indeed, normally implied in the
word.

Terms like 'development', 'rationalization', and 'industrialization' (which
likewise are occasionally used loosely to cover the whole wider process) also
will better be restricted to their more precise meanings — that is, respec-
tively, technical development of whatever degree, technical rationalization,
and the preponderance, over other sectors of the economy, of technically
developed industry. The first two terms, technical 'development' in some

52 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

degree, and 'rationalization', can then usefully refer not only to phases
within technicalism but even to isolated situations of a very pre-Modern
date, not organically related to the complex of Modern times at all. As to
'Westernization', the broad use of such a term implies a parti pris on the
question — crucially open for us— of how much the complex of traits in
question may be dissociable from the rest of Western culture, in particular
from the pre-modern Occidental cultural heritage. I prefer to restrict the
term to an explicit adoption of Western traits as Western rather than simply
as Modern.

Technicalistic, Technical Age

I prefer to use special variants on the root 'technical' when it is important
to be precise. The term 'technicalized', corresponding to 'industrialized'
but applicable to technical development not only in manufacturing but
in agriculture, administration, science, and so on, will suffer neither semantic
loss nor serious obscurity if it is used to characterize the several sectors
or the whole of a society in which the dominant elements are on a level of
social organization where, in intellectual and practical activity, calculatively
rationalized and specialized technical procedures form an interdependent and
preponderant pattern. Denmark, which may not be primarily industrialized
as a society, is highly technicalized in this sense. Then I shall use the word
'technicalistic' to refer to patterns of thought or activity appropriate to or
functionally associated with technicalized processes.

Just as the term 'agrarianistic' cannot safely be used of all developments
which characterized the agrarianistic societies, so the terms 'technicaliza-
tion' and 'technicalism' will not exhaust all the traits legitimately associable
with Modern times. Above all, certain moral qualities which may have been
necessary to launch technicalization in the first place, or the cultivation of
which may be facilitated by its presence and consequences, cannot be
subsumed under 'technicalism'. I prefer cautiously to refer to them as
typical of or associated with the age of technicalism, the period since tech-
nicalistic patterns became crucially dominant in the world at large. The
period thus comes after the generation during which technicalization came
into full effect in at least some aspects of some west European societies
with the Industrial (1785) and the French (1789) Revolutions, which was also
the generation of the establishment of European world hegemony. For
convenience I call this succeeding age (down at least to the present) the
'Technical Age', since an age can be named merely for a dominant feature
and no one is likely to find the term misleading in this brief form. As a term
for an age it refers to the time-period the world over, whether any given
land was actually being technicalized then or was merely suffering {or even
fortunately escaping) the backwash effects of technicalization elsewhere.
Correspondingly, when I use the term pre-Technical, it refers generally to

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 53

the time-periods that precede the Technical Age; it does not refer to non-
technicalized, or technically underdeveloped, countries or sectors within the
Technical Age. This allows me, as the less precisely used 'Modern Age' and
'pre-Modern' would not, to keep in focus the overall historical situation as
well as the internal state of development in any given place. It will be
obvious that the terminus a quo of the Technical Age is at the same time the
terminus ad quern of the Agrarian Age.

West, Occident, Europe

I use the term 'West' in discussing conditions of the Technical Age; I do
not use it before. I must distinguish it from my use of the terms 'Europe'
and 'Occident'.

The term 'West' is often very loosely used, even if not quite so misleading
as the term 'East'. It can cover at least five different historical groupings
of lands and peoples. All too often a statement proper to one of these group-
ings is taken implicitly as applying to another, with consequent confusion.
(The confusion usually occurs in tendentious ways.) 'West' refers (i)
originally and properly to the western or Latin-using half of the Roman
empire; that is, to the west Mediterranean lands. By extension, (2) it can
refer to the west European lands generally, more precisely to the western
or Latin Christian countries north of the Mediterranean (in the Middle Ages
and since) which traced their heritage back to the Roman empire. In this
usage it will normally exclude those west Mediterranean lands which turned
Muslim. By a further extension, (3) the overseas settlements of the west
European peoples anywhere, in the Americas and the Antipodes, can be
included. It is in these latter two senses that the term 'Occident' is used
in this work : to designate the peoples of Western (originally Latin) Christen-
dom, including in more modern times their overseas settlements. (The
term 'Occident' is less popularly used in English and so has some chance
of being captured for a precise purpose.) The term 'West' is also sometimes
used (4) for all European Christendom, usually including all the peoples of
European origin (both in the European region and in their extra-European
settlements); that is, both the original west Europeans and the east
Europeans from whom the term 'west' originally distinguished them. It is
by a curious extension of this usage that selected portions of ancient Greek
history are commonly included (retroactively) in a 'Western' history which
otherwise is chiefly only Occidental in the narrower sense (so that the
Merovingians and not the Byzantines are made to appear as 'Western',
with the consequent implication of being heirs of ancient Greece!). For
such purposes, the terms 'European' or 'Christian European' are more
appropriate, at least for the period before the Technical Age, when extra-
European settlers were still obviously Europeans. But for the strictly
Modern period, I use the term 'Western' for this purpose so as to include

54 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

unequivocally the extra-European settlers. Finally, (5) sometimes all
Afro-Eurasian civilized lands west of the Indus may be included, that is,
roughly both Europe and the Middle East; for such a purpose, 'west
Eurasian' or 'Irano-Mediterranean' is preferable.

The term 'Europe', in contrast to the term 'West', has come to have a
specious exactitude, referring to the most westerly peninsula of Eurasia,
with the associated islands and an arbitrarily designated part of their
continental hinterland, usually bounded at the Urals. This precise area,
however, has at no time ever formed a cultural or political entity to any
degree at all. I have used the vaguer term 'European region' or 'European
lands' to cover an area similar to this but forming at least in pre-Modern
periods a historically truer grouping: the lands north of the Mediterranean
(from Anatolia to Spain) with their hinterland northward (including into
the Russian plains), and without always excluding such related lands as
the Maghrib. The term 'Europe' then can take on a more usefully precise
political and social meaning only in Modern times, when it refers to the
west European stages plus those east European states that had assimilated
to them; i.e., until recently, the Christian states.

Accordingly, in Book Six, I still use 'Occident' to refer to the ex-Latin
Europeans and their overseas settlers, and 'European' in the general regional
sense of the lands north of the Mediterranean. But I also use 'Western'
to refer to all European Christendom and its extra-European extensions,
and 'Europe', in a political sense, to refer to the Christian European states
and their organizations.

Yet another recent usage of 'West' comes back to a west-east division
within Europe, but rather different from the old one : the 'East' is the Com-
munist bloc, and the 'West' then becomes the non-Communist Western bloc.
In a general historical context it will be easy to refer to 'Western-bloc'
powers when this is intended.

'East Roman empire' , 'Byzantine'

After the capital of the Roman empire was finally settled in Constantinople
(following a period when Rome, in any case, had not been the actual capital),
many authors call the empire 'East Roman' or 'Byzantine'. The term
'Byzantine' becomes useful at least to distinguish the ruling classes, though
it is most appropriate only after the Arab conquests, when Latin practically
disappeared. But the term 'East Roman' can be misleading in a more general
historical perspective. It does not happen to refer, like 'Eastern Han', to
the shift of the capital from a western to an eastern location; rather it
refers to the frequent appointment, during more than a century, until 480,
of an autonomous co-emperor in the western provinces. This used to be
misconstrued as marking the creation of two contemporaneous empires, an
'East Roman' and a 'West Roman'. The 'West Roman' was then tacitly

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 55

identified with the Roman empire proper — partly because it included the
original Roman territories, but more because the historians, as west
Europeans, were chiefly interested in the western provinces.

Accordingly, when during the fifth century most of the western provinces
temporarily escaped imperial control, never to be fully reconstituted as a
group, this was thought of as the 'fall of the Roman empire' (dated speci-
fically in 476, a date poorly chosen in any case). And, in fact, Gaul and
Britain, on whose history most Western historians have tended to centre, were
not reoccupied; so the imperial power does end locally there in the fifth
century. Thereafter by such scholars the continuing main body of the Roman
empire has been called the 'East Roman empire' and has been thought of as
distinct from the 'Roman empire' proper; so that even the re-establishment
of imperial authority in the more important of the western provinces in the
sixth century has sometimes been thought of not as a restoration but as an
expansion of a different empire — 'the East Roman empire at its largest
extent'. Commonly used historical atlases so label their maps. But we must
be continually reminded that Gaul and Britain were marginal to the empire
as such, and the whole western half of the empire was in most ways less
important than the eastern half. From the viewpoint of the society and
culture of the empire as a whole, there was no beginning 'division' into east
and west, and the major transitions come not in the fifth but in the third and
the seventh centuries. To refer to the later undivided empire as 'East Roman'
is to retroject Occidental independence into too early a past and to obscure
the continuity of the empire as a whole. It can confuse our sense of the
Roman heritage in the Mediterranean basin as it confronted the first Muslims,
as well as distort our sense of the early relations between Islamdom and the
Occident.

The Westernizing world-map projections and atlases

The Muslims in ages past had an image of the world noticeably more balanced
than that of the west Europeans. Maps in certain Western history books
labelled 'the known world' represent, of course, not the world as known in
more advanced and cosmopolitan centres, but as known to the literate
public of western Europe. Given the peripheral location of the European
peninsula, the Occidentals' image of the world might be expected to have
been eccentric. Their division of the old world was not into seven parts, as
was the Muslims', but into only three, centred on the Mediterranean Sea
(the lands north of it were Europe ; those south of it, Africa ; those east of it,
Asia). Such a distribution was naturally totally inappropriate to the hemi-
sphere as a whole.

Yet it did serve admirably to set off the little European peninsula as a
unit comparable to the great land masses. The Medieval ethnocentric classi-
fication was preserved and subsequently written into the modern Western

56 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

map of the world, just as was the equally ethnocentric conception 'Orient'.
Its absurdity was disguised by the increasingly widespread use of a drastically
visually distortive world map, the Mercator projection, which by exaggerat-
ing northward manages to make an artifically bounded 'Europe' look larger
than all 'Africa', and quite dwarf that other Eurasian peninsula, India. In
this way all the 'well-known' cities of Europe can be included, while the un-
familiar cities of India can be omitted. When the Mercator projection is
decried for distorting Greenland — as if Greenland mattered so much — map-
makers can resort to a remarkable compromise: a projection like Van der
Grinten's preserves the basic ethnocentric distortion, continuing to exagger-
ate Europe at the expense of the other main historical cultural centres, but
without exaggerating so much the barren far north. Even in historical atlases,
maps of the world tend to be visually distortive, and they almost invariably
place the Atlantic and Europe in the centre. In our day, such maps, putting
the White men's lands on so much larger a scale and in so much more pro-
minent a position than the Coloured men's lands, may fittingly be called 'Jim-
Crow' world maps. But the idea is much older than modern racialism. Such
maps represent visually how pre-Modern yet continuing popular notions can
persist into even Modern scholarly usage.

The disastrous effect that distortive map images can have has been
recognized in modern military thinking, where the Second World War forced
at least airmen to use 'global' maps so as to think imaginatively. Unfor-
tunately, many Islamicists, like other scholars, have unconsciously allowed
their physical image of Islamdom to be distorted by the popular maps that
are all about them. They continue to be influenced by European-centred
ways. Yet Islamicists, at least, could very usefully take their geographical
terms and their 'world image,' and even their 'maps of the known world'
from pre-Modern Islamdom. An atlas so oriented would be of great value in
reinforcing the proper views presented in carefully written works.

USAGE IN ISLAMICS STUDIES

'Islamics', 'Islamicist'

When the object of a body of studies is people who themselves make studies,
it can on occasion be useful to distinguish, in terms, between the studies and
their object, which is not done in phrases like 'Islamic studies', 'Islamic
scholar', 'Chinese studies', 'Chinese scholar', equally used for studies about
or by Muslims or Chinese. Hence the use of slightly pretentious terms like
'Sinology', 'Sinologist', and Tslamology' or 'Islamics', and 'Islamicist'.
Though I use the term 'Islamics' I feel it does not yet distinguish clearly
enough between studies of Islam as such and studies of Islamdom. The
hospitality of English for appositional phrases permits constructions at once
simpler and more unmistakable: 'China studies', 'China scholars', 'Islam
studies', 'Islamdom studies', 'Islamdom scholars' — a form adaptable at will

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 57

to any desired delimitation of field. But despite the self-evident clarity of such
phrases, I hesitate to introduce them — at least before the term 'Islamdom'
has established itself.

'Islamdom' , 'Islamicate'

The use of the unwonted terms 'Islamdom' and 'Islamicate' requires a
more extended defence. I plead that it has been all too common, in modern
scholarship, to use the terms 'Islam 'and 'Islamic' too casually both for what
we may call religion and for the overall society and culture associated
historically with the religion. I grant that it is not possible nor, perhaps, even
desirable to draw too sharp a line here, for (and not only in Islam) to separate
out religion from the rest of life is partly to falsify it. Nevertheless, the
society and culture called 'Islamic' in the second sense are not necessarily
'Islamic' in the first. Not only have the groups of people involved in the two
cases not always been co-extensive (the culture has not been simply a 'Muslim
culture', a culture of Muslims) — much of what even Muslims have done as a
part of the 'Islamic' civilization can only be characterized as 'un-Islamic' in
the first, the religious sense of the word. One can speak of 'Islamic literature',
of 'Islamic art', of 'Islamic philosophy', even of 'Islamic despotism', but in
such a sequence one is speaking less and less of something that expresses
Islam as a faith.

Accordingly, it should avoid confusion, to distinguish the two current
meanings of 'Islam' conceptually by means of distinct terms. If one speaks of
'Islamic law', for instance, one may mean the Shari'ah; but if one is compar-
ing law, as a dimension of cultural life, with 'Islamic' art or literature, one
should include the non-Shar'i legal patterns on a level with the 'non-religious'
literature and art. Otherwise one gets a false balance. But too rarely is the
non-Shar'i law, in fact, included. Without a distinction of terms, such a con-
fusion, which may originate in the chances of what material is available to
scholarship, tends to persist unnoticed. Thus there are several studies of
'Islamic international law' which (taking 'Islamic' in the first, narrower,
sense) deal with the Shar'l principles of the relation of the caliphate to non-
Muslim states. But it seems to have occurred to few to study that other
'Islamic international law', not explicitly religious but characterizing the
civilization as such, a law which governed, above all, relations among
'Islamic' states. Yet the latter, and not the Shar'i 'international' law, is what
would culturally correspond to most of what we call 'Islamic art', or 'Islamic
literature', or 'Islamic science'; and, indeed, to Western 'international law'.

In fact, the need for a distinction is rather urgent. It would be easy to show
that not only beginners but even scholars have found themselves falling into
outright error because they have not kept the two current senses of the word
'Islam' distinct. A study of 'Medieval Islam' or of 'Modern Islam' may be
primarily a study of religion, or it may be a study of an overall culture in

58 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

which religion simply takes its place ; or it may be a mixture, sections of it
differing according to different sources of information. It has happened, for
instance, that the same discussion referred to 'Medieval Islam' in a broader
cultural sense and to 'Modern Islam' in a more specifically religious sense,
and that the fact went unobserved that different discussants, or the same
discussant at different times, were referring to different matters in the two
cases. The results can be most misleading. Bernard Lewis has suggested that
the adjective 'Islamic' be used in the second, the cultural sense, and the
adjective 'Muslim' in the first, the religious sense. But it does not appear that
this usage will be maintained; and indeed there is some advantage in distin-
guishing between 'Islamic' as an adjective 'of or pertaining to' Islam as either
an idealized or a historical cumulative tradition of faith, and 'Muslim' as an
adjective 'of or pertaining to' the Muslims, insofar as they accept that faith —
a slight distinction, but sometimes a useful one, and one that comes easily.

I have come to the conclusion that the problem can be solved only by
introducing new terms. The term Tslamdom' will be immediately intelligible
by analogy with 'Christendom'. Tslamdom', then, is the society in which the
Muslims and their faith are recognized as prevalent and socially dominant,
in one sense or another — a society in which, of course, non-Muslims have
always formed an integral, if subordinate, element, as have Jews in Christen-
dom. It does not refer to an area as such, but to a complex of social relations,
which, to be sure, is territorially more or less well-defined. It does not, then,
duplicate the essentially juridical and territorial term, 'Dar al-Islam' ; yet, in
contrast to 'Muslim lands', it is clearly collective — frequently an important
point. Sometimes the phrase 'the Islamic world' is used much in this sense.
I prefer not to use it for three reasons: [a) in compound phrases where Tslam-
dom' can be a useful element, the three-word phrase can become clumsy;
(6) the phrase itself uses the term 'Islamic' in too broad a sense ; (c) it is time
we realized there is only 'one world' even in history. If there is to be an
'Islamic world', this can be only in the future.

On the other hand, if the analogy with 'Christendom' is held to, Tslamdom'
does not designate in itself a 'civilization', a specific culture, but only the
society that carries that culture. There has been, however, a culture, centred
on a lettered tradition, which has been historically distinctive of Islamdom
the society, and which has been naturally shared in by both Muslims and non-
Muslims who participate at all fully in the society of Islamdom. For this, I
have used the adjective Tslamicate'. I thus restrict the term 'Islam' to the
religion of the Muslims, not using that term for the far more general pheno-
mena, the society of Islamdom and its Islamicate cultural traditions.

The noun Tslamdom' will presumably raise little objection, even if it is
little adopted. (I hope, if it is used, it will be used for the milieu of a whole
society and not simply for the body of all Muslims, for the Ummah.) At any
rate, it is already felt improper, among careful speakers, to refer to some
local event as taking place 'in Islam', or to a traveller as going 'to Islam', as

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 59

if Islam were a country. The adjective 'Islamic', correspondingly, must be
restricted to 'of or pertaining to' Islam in the proper, the religious, sense, and
of this it will be harder to persuade some. When I speak of 'Islamic literature'
I am referring only to more or less 'religious' literature, not to secular wine
songs, just as when one speaks of Christian literature one does not refer to all
the literature produced in Christendom. When I speak of 'Islamic art* I
imply some sort of distinction between the architecture of mosques on the
one hand, and the miniatures illustrating a medical handbook on the other —
even though there is admittedly no sharp boundary between. Unfortunately,
there seems to be no adjective in use for the excluded sense — 'of or pertaining
to' the society and culture of Islamdom. In the case of Western Christendom
we have the convenient adjective 'Occidental' (or 'Western' — though this
latter term, especially, is too often misused in a vaguely extended sense).
'Occidental' has just the necessary traits that 'West Christian' would
exclude. I have been driven to invent a term, Tslamicate'. It has a double
adjectival ending on the analogy of 'Italianate', 'in the Italian style', which
refers not to Italy itself directly, not to just whatever is to be called properly
Italian, but to something associated typically with Italian style and with the
Italian manner. One speaks of Ttalianate' architecture even in England or
Turkey. Rather similarly (though I shift the relation a bit), 'Islamicate'
would refer not directly to the religion, Islam, itself, but to the social and
cultural complex historically associated with Islam and the Muslims, both
among Muslims themselves and even when found among non-Muslims.

The pattern of such a double adjectival ending, setting the reference at
two removes from the point referred to, is sufficiently uncommon to make me
hesitate. But there seems no alternative. In some contexts, but only in some,
one can refer without ambiguity to the 'Perso-Arabic' tradition to indicate
'of or pertaining to' Islamdom and its culture, for all the lettered traditions
of Islamdom have been grounded in the Arabic or the Persian or both. In
other cases, one might use a periphrasis involving the terms 'traditions/
culture/ society of Islamdom'. One cannot, speaking generally, call Swedish
'a Christian language' ; and if one were debarred from calling it an 'Occidental'
language, one could not say simply that it is 'a language of Christendom',
which might in some contexts seem to imply that it was to at least some
extent used throughout that extensive realm; but one might say it is 'a
language of the culture of Christendom'. Likewise, it is hardly accurate,
despite certain West Pakistani claims, to call Urdu an 'Islamic' language, in
the strict sense. (It was the insistence of some Muslims on treating it that
way, and opening a meeting on fostering Urdu with Qur'an readings, that
drove Urdu-loving Hindus away from it and may, in the end, have meant
the ruin of Urdu in its motherland.) If one could not refer to it as Tslamicate',
one could yet say it was a language 'of the culture of Islamdom'. One cannot
refer to Maimonides as an Islamic philosopher, but one could say, without
being seriously misleading, that he was a philosopher in the Perso-Arabic

60 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

tradition or, still better, a writer in the philosophical tradition of Islamdom.
But there is a limit to such periphrases. Eventually, it is stylistically less
clumsy to use an explicit term. Moreover, such a term may have valuable
pedagogical uses, its very presence militating against the confusions which
periphrases would avoid in the writer but not necessarily in his readers.

It may be noted that some, not only Arabs and Western Arabists, but
latterly even some non-Arab Muslims (for the historical reasons noted else-
where), might use the term Arabic' — especially in such a case as that of
Maimonides. But — to take the case of philosophy — this is ruled out because,
for one thing, some important representatives of that tradition wrote in
Persian. In fields other than scholarship and philosophy — in politics or art,
say — the idea becomes even more patently absurd, despite the bias in
favour of it among certain scholars. The term 'Arabic' must be reserved for
that subculture, within the wider society of Islamdom, in which Arabic was
the normal language of literacy; or even, sometimes, to the yet smaller
sphere in which Arabic-derived dialects were spoken. Indeed, the Western
temptation to use this term with a wider reference springs from historical
accidents that have tended falsely to identify Arab' and 'Muslim' in any
case. To use the term 'Arabic' then, would not only be inaccurate, it would be
one of those erroneous usages that reinforce false preconceptions — by far
the most mischievous sort of error, as I have noted in the section on historical
method above.

'Middle East', 'Nile to Oxus'

For this region I will not usually use the term 'Middle East' but one or
another phrase in 'Nile to Oxus' — -'from Nile to Oxus' (generally implying
inclusively), for instance, or 'between Nile and Oxus' if what is meant is
some given spot in or portion of the area; or occasionally 'in the Nile-to-
Oxus region'. The term 'Middle East', which seems the best phrase of those
more commonly used, has a number of disadvantages. It is, of course, vague,
being used for so limited a region as that of the eastern (or even northeastern)
Arab lands plus Israel (the presence of the latter is probably the occasion for
having recourse to such a vague term for so limited an area) ; and it has been
used for so extensive a region as that stretching from Morocco to Pakistan,
and sometimes including a number of Muslim peoples even further afield. It
can, of course, be defined at will ; but overtones remain, especially overtones
implying an Iran of present-day political bounds.

Its principal disadvantage stems from its relatively exact military usage,
where it originated. It cuts the Iranian highlands in half — the western half
('Persia') having been assigned to the Mediterranean command, the eastern
half ('Afghanistan') to the Indian command. Since the Iranian highlands are
of primary importance in the region that is basic to Irano-Semitic and
Islamicate history, such a usage is completely unacceptable. Unfortunately,

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 6l

the military usage as to the eastern limits of coverage has become standard
in a great many works using the phrase 'Middle East', and for many readers
it comes to imply an area that is, on balance, more westerly than our history
requires. Since for Westerners there is anyway a temptation to see everything
from the shores of the Mediterranean, the more easterly parts becoming
foreshortened as it were, a phrase that has the virtue of explicitly running
counter to that temptation is to be preferred.

The phrase 'Nile to Oxus' has the further virtue of being eminently con-
crete: if one means 'Egypt', say, or 'Egypt and Syria', or some other rela-
tively restricted area (as many do in fact, even when they believe they are
using the term 'Middle East' in an inclusive sense) one may hesitate to make
too sweeping a generalization if forced to ascribe it explicitly to all the lands
from Nile to Oxus.

Another disadvantage of the term 'Middle East' is that it implies it is part
of some 'East' — that is, all civilized lands but the Occidental, taken as
somehow forming a civilization or a region to which something distinctive
may be ascribed, set off as one entity against the 'West' as another. That
sort of Western ethnocentrism is discussed in the preceding section on
usage in historical studies. The same objection applies to the term 'Near
East', which has the further disadvantage of shifting the focus to a yet more
westerly zone than 'Middle East' usually does. The absurd phrase 'Near and
Middle East', a compromise sometimes used, has the disadvantages of both
phrases and the advantages of neither.

From the point of view of the Oikoumene as a whole, 'Middle West'
might do, but would anyway not be so immediately intelligible as 'Nile to
Oxus'. The phrase 'West Asia' (which seems gratuitously to exclude Egypt
from a region where historically it commonly belongs if it belongs to any
region beyond itself) has the disadvantage of perpetuating the notion of
'Asia' as a 'continent', a notion that is merely a variant on the Western
ethnocentrism of the term 'East'. The Germans commonly use the term
'Orient' for the area from Nile to Oxus. This usage might be both unex-
ceptionable and convenient if only that word did not have far different
connotations for English-speaking readers — and also among Germans, to
judge by the frequency with which German writers use it in both a strict
and an extended sense even in the same discussion.

It will be noted that western Anatolia does not lie between Nile and Oxus,
but lies on the contrary along the northern shores of the Mediterranean.
This has been its location in all periods, from the time of the ancient Lydians
who patronized Delphi to the time of Atatiirk.

For the most important complex of cultural traditions in the region from
Nile to Oxus, I use the phrase Trano-Semitic', which refers to the cultural
traditions, both on the folk level and on that of the high culture, rather than
to the area as such. The two terms have only for a brief time, if ever, been
geographically coextensive, for the area in which the Irano-Semitic culture

62 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

prevailed in its various periods steadily expanded. For an explanation of my
usage of that phrase, see note 9 in chapter I of Book One.

'Arab'

The term 'Arab', 'arab, as noun and adjective, has been used on at least
five levels. (1) It has referred — perhaps- originally — expressly to the Bedouin,
the nomads and especially the camel nomads of Arabia. (But careful usage
has preferred a special term, a'rab, for them.) -This has at times been the
commonest usage of the term in Arabic. However, to render the Arabic 'arab
in such cases by our 'Arab', as some writers do when translating, is bound to
be confusing and is to be avoided. The reader must be on the alert for such a
usage in older translations. When a pre-Modern Arabic writer, such as Ibn-
Khaldun, said something uncomplimentary about 'arabs, he was usually
speaking only of the Bedouin. (2) Then it has referred to all those claiming
descent from or old cultural identification with the Bedouin or their language,
including of course the 'settled Bedouin'. In this use, it has historically
sometimes had an implication of 'Muslim', since the early Muslims were
Arab in rather this sense; but the early jurist Abu-Yusuf, using it in this
sense, included also some Christians and Jews. (3) The next extension was to
all those peoples speaking Arabic-derived dialects", whatever their relations
to Bedouin traditions or to Islam; in this sense, whole peasant populations
can be called Arabs. (However, those among whom the literate have used
some other than the standard Arabic alphabet — for instance the Maltese
and some other non-Muslim groups, especially Jews — have commonly, but
not always, been excluded.) This latter- sense is essentially a modern one. In
using it in this work I am retrojecting it, for convenience' sake, upon a set of
groups which might not have recognized that they formed a common
category; it is analogous to having a common term, 'Latins', for all the
Romance-speaking peoples. It must not, therefore, be lent any 'national'
overtones: in this sense, 'the Arabs' have moved toward forming a nation
only recently. However, it is the commonest modern usage and it must be
remembered that in this sense the Arabs are mostly neither Bedouin nor
tribal; they are, in large majority, peasants, living in villages and closely
tied to the land. (4) It has further been used where the normal language of
literacy was the classical Arabian, or Mudari, Arabic — whether the home
vernacular was Arabic-derived or quite unrelated. Usually this usage has
been restricted to the individual level. Persons who wrote in Arabic but
whose own language was Persian or Spanish or Turkish or Kurdish have
been included in what is called collectively 'the Arabs'. Where whole peoples
have possessed literacy only in Arabic — e.g., the Somalis — this usage has not
usually applied. But even at best the usage is very dubious ; it is sounder to
say something like 'the Arabic writers'. (5) Finally, there are to be found
authors who will seem to use the term for all peoples among whom Arabic

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 63

is used at least in ritual. This is never done consistently ; but it seems to be
the implied definition when a book on the modern Arabs, for instance, in-
cludes, as illustrations of past Arab achievements, pictures of the Taj
Mahall in India or of illustrations to Persian poetry. Such a usage is thoroughly
confusing and unacceptable.

'Allah'

We properly use the same English word for the object of worship of all who
are recognized as monotheists — the various sorts of Christians and Jews, as
well as many persons of other faiths. Normally we leave untranslated a
proper name, such as 'Zeus' or 'Odin', when we think of the divinity in
question as distinct from (and lesser than) the monotheists' One God. To
use 'Allah' in English can therefore imply, accordingly, the notion that
Muslims hdnour something different from what is honoured by Christians and
Jews (and Stoics and Platonists and so forth), and presumably something
imaginary: as if they believed that it was some mythical god called 'Allah',
rather than God, the Creator. This is essentially a dogmatic position and can
be allowed only in those ready to admit its theological implications. Some-
times Muslims writing in English use 'Allah' instead of 'God', generally with
a like distinction in mind, but with the implications reversed — implying
more or less consciously that the 'God' which Christians and Jews worship
is not really true in the full sense, so that Muslims must be distinguished as
worshipping 'Allah'. Sometimes, to be sure, the usage merely reflects an
understandable zeal for the Arabic text of the Qur'an.

On the other hand, in philological and historical contexts it is sometimes
useful to distinguish a particular figure with lineaments envisaged by
particular groups in particular forms. In this case, just as it is sometimes
convenient to refer to 'Yahweh' in discussing the early Hebrew conception
of God, so it can be convenient to refer to 'Allah' in corresponding circum-
stances.

'Hadith report', 'Tradition'

The term hadith has often been translated 'tradition', in the sense of the
Latin traditio, something handed from one to another, used of certain
alleged unwritten laws and teachings in Jewish and Christian theology.
When scholars were Latinists and theologians and when it was considered
proper to interpret an alien culture in terms of one's own concepts, the
rendering 'tradition' was doubtless convenient. Now, however, it is felt to be
important to understand a culture in its own terms. In English, the word
'tradition' implies not only a contrast to anything written, but anonymity
and imprecision. The hadith reports, however, are not a matter of vague
custom but of explicit statements, texts, early put into writing; frequently
just contrary to custom; and always naming both the transmitters and the

64 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

original source. Since there is also, of course, tradition among Muslims, in
the English sense of the word — and since that tradition is often to be con-
trasted to hadith — such a term as 'narration' or 'report' seems a far more
convenient rendering for hadith if the term is to be translated at all. Hence
in this work I refer to hadith reports and hadith reporters (or transmitters)
where conventionally writers have referred to 'Traditions' and Traditionists' ;
I refer to the hadith corpus where conventional writers have referred to
'Islamic Tradition'.

A term like 'report' is also philo logically more accurate, for hadith means
'new' and then reported 'news', 'narration' (and finally even 'conversation').
Moreover, theologically, the analogy with Christian 'tradition' is technically
unsound. In exact usage, the term hadith, 'report', is explicitly distinguished
from sunnah, sunan, 'custom(s)', and from ijmd, 'consensus'; it is a report of
sunnah. It is the term sunnah that would be more properly compared to the
Christian 'tradition' — at least sunnah when supported by ijma', consensus,
whether the sunnah happens to be supported by hadith reports or not.

But what matters is not so much the philological accuracy of the rendering
as its scholarly consequences. By using the term 'report', I have left the
words 'tradition' and 'traditional' open for more appropriate uses — which will
sometimes correspond to sunnah in its more general sense ('living' tradition,
in the works of Joseph Schacht), but will more often refer to still other cases
of the basic cultural phenomenon described in the Prologue. Thus I can
avoid giving occasion to certain confusions that are too common in the
scholarly literature as it is. Too many writers seem to suppose that Muslim
traditional lore and hadith lore, Muslim traditional law and Shar'i law, were
the same, or would have been the same if the Muslims in question had been
'true' Muslims. Or at least many writers suppose that hadith lore and the law
based on it were always 'traditional' in some ordinary English sense, even
if they did not form the whole tradition. Often writers will use 'tradition' and
'traditional' freely in their ordinary sense in the same discussion where they
use these words for hadith reports, without specially marking the contrast.
But such ways of thinking and writing make nonsense of all the Shar'i
reformers from al-Shafi'i to Abd-al-Wahhab. But these men were all attack-
ing actual Muslim custom and tradition in the name of (often obscure) texts,
nass, which were commonly in fact innovative. They could claim to represent
tradition, in the crucial cases, only in a very tenuous sense even if one grants
them all their presuppositions: that is, in the sense that those few who
followed a lone but exalted precedent, contrary to all custom, were main-
taining the only tradition that really counted. In fact, the hadith reporter
was not, as such, a traditionalist but a textualist (or perhaps by intention a
revivalist — in the exact sense of that term) and was often quite as much
opposed to the truly tradition-minded as to any rationalist.

The needless confusion invited by using the term 'Tradition' for a report
has been seen by a number of scholars, who have tried to avoid it by capita-

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 65

lizing the word when used in its 'technical' sense. This might work if it were
not that the confusion is weighted; the usage is one of those insidious mis-
usages which reinforce erroneous tendencies already present among us, and
which therefore cannot be counteracted with a mere footnote warning. For
unfortunately the hadith-minded Muslims themselves insisted that the
hadith reports did represent the only tradition that ought to be valid for a
Muslim, and moreover a tradition unbroken from the time of the Prophet
himself. Until the work of Ignaz Goldziher, Islamicists tended to believe
this without examination; even since his time they have been tempted, as
philologians, to look on the hadith textualist party among Muslims — the
party most easily traceable by philological methods — as in all cases and in
all periods the partisans of the 'true' Muslim tradition, in the ordinary sense
of that word, even if they acknowledged that some 'traditions' were in fact
later than was claimed. This has led them to overlook the anti-traditionalist
function of hadith reports in forming the early law, and to think of Hanbalis
and Zahiris, who clung especially to hadith reports, miscalled 'traditions',
as the most traditionalist — not merely the most textualist — of the schools: a
serious misconception. And it has led them, if at all alert, to having to have
recourse to needlessly cumbersome excursuses to explain, apropos of Muslim
Modernist movements, that hadith reports ('Tradition') form only a part of
the actual Islamic tradition; that a stress on such reports may or may not
indicate what is ordinarily called traditionalism in a man; that, in fact, a
strict hadith-mindedness, now as ever, may well imply an anti-traditional
orientation in several respects.

Practically any discussion involving the role of hadith reports will illu-
strate the potential and often actual confusion that results from using the
rendering 'Traditions'. Too often not only the unwary general reader but the
scholar himself falls victim. It is true that Islamicists, like other scholars, tend
to have their own traditionalism (in the common sense), but scholars know
that when a usage is not merely unfortunate but actively misleading, it must
be abandoned, however painful the effort to do so and however hallowed the
usage is by age.

An example may serve to pinpoint the difficulty. George Makdisi, in an
article cited in chapter III of Book Three, shows that the conventional
picture of Ash'arism as becoming 'orthodox' in the Earlier Middle Period
will not stand up. But he does not escape the conventional picture otherwise,
for he identifies traditionalism among early Muslims with adherence to
hadith reports; and then he finds himself puzzled by his own results in
consequence. This picture of Islamic history was almost certainly reinforced
by his use of the term 'Tradition' for a hadith report; at any rate, his
terminology did nothing to alert him to the confusion.

He begins with a remark that it would be only natural — on the basis of
general history of religions — to expect (as in the usual scholarly image of
Islamic development, based on the Ash'ari apologists) an early division be-

66 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

tween 'traditionalists' and 'rationalists', which would then be later bridged
by a mediating 'orthodoxy' (the Ash'ari kalam). In saying this, he seems to
identify the 'traditionalists' in the general sense with his 'Traditionists', the
hadith-reporters and especially the Ahl al-Hadith — and, of course, the
'rationalists' with the Mu'tazilis (as against the Ash'aris).

But, in fact, since both the 'kalam men' and the 'hadith men' had at first
been, in different ways, partly opposed to the living tradition of Marwani
times, to call the hadith men 'traditionalist' in any broad sense is to mis-
represent them (though, to be sure, it was their own misrepresentation) ; the
Mu'tazilis had as good a claim to be called traditionalist, their 'rationalism'
(in defence of the older tradition) being no more anti-traditional than was the
textualism of the hadith men (or than the 'rationalism' of the Ash'aris). To
the extent that the hadith men succeeded in gaining popular support, their
subsequent conflict with the kalam men (Mu'tazili or Ash'ari) was a
secondary development not to be expected to answer to any universal
primitive experience in religious traditions.

Indeed, Makdisi goes on to note that in Islam the pattern he expected to
see did not fulfill itself after all. But such an observation loses its point if one
gets outside the framework set by the identification of hadith reports with
Islamic tradition. Without such an identification, he might have been led to
see that what he had to deal with were three, not two, universal types of
phenomena in religious history — traditionalism, 'rationalism', and textualism
(allowing, but only for the moment, the legitimacy of a term like 'rationalism'
for the argumentative viewpoint of the kalam men). He then might have
seen the whole range of Islamic textualism in a different light — which might
not have altered his immediate conclusions, but would surely have given
them an ultimately different and more fruitful meaning. Avoidance of the
term 'Tradition' for a hadith report and of 'Traditionist' for a reporter
would not by itself have led to new insights, but it might have helped avoid
taking old misconceptions too much for granted.

'Sect', 'firqah'

The common term used in the the pre-Modern Arabic and Persian languages
for any grouping of people according to their opinions was firqah. This has
been translated 'sect', but it rather rarely answers to the modern English
notion of 'sect'. Usually it should be rendered by nothing stronger than
'school of thought'. Often it is used to refer to a single teacher and his
disciples, with reference to one minor point of doctrine. In such a case there
is no question of a body of persons sharing a common religious allegiance
such that their overall religious life is led among themselves and apart from
others, as is implied in a 'sect'.

The Muslim historians of doctrine always tried to show that all other
schools of thought than their own were not only false but, if possible, less

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 67

than truly Muslim. Their works describe innumerable 'firqahs' in terms
which readily misled modern scholars into supposing they were referring to
so many 'heretical sects'. These histories of doctrine have been called, not
unfairly, 'heresiographies', which, however, is rather a description of their
tendencies than an exact designation of their contents. But to use the word
'sect' wherever a Muslim writer used the word firqah produces odd mis-
conceptions. A person could maintain a given viewpoint on the imamate,
one on questions of metaphysics or kalam, and one on fiqh law ; he could be,
for instance, a Jama'i-Sunni, a Mu'tazili, and Hanafi. Beginners, and (un-
fortunately) not only beginners, have sometimes been confused as to how so
many Hanafis, whom he knows to be Sunnis, could belong to the supposedly
rival 'sect' of the Mu'tazilah. In so prominent a case the confusion can be
relatively easily laid to rest; but the elevation of many less well-known
viewpoints on one or another issue into full-blown 'sects' has peopled with
strange ghosts the history of Muslim rebellions and urban factions as well
as the history of doctrine.

Note that the conventional distinction between 'orthodox' Islam and 'the
sects' is at best dubious. For my usage of the terms 'Sunni' and ' Jama'i', see
the discussion in chapter I of Book Two.

NOTE ON TRANSLATING

A translation must be judged according to the purposes of the given trans-
lator. It is possible to distinguish three usual sorts of translation : re-creative,
explanatory, and precise study translations. In bibliographical references I
have identified translations accordingly, when necessary.

The translator's purpose may be to re-create a work so as to enrich a
second literature with a work having approximately the same effect as the
original. Sometimes the effect intended is very close to the original effect,
sometimes it is an analogous effect. One may then draw inspiration from
such a model without being tied to those features of it that are relevant only
within its original setting. Such a translation of poetry will aim to be poetic ;
with prose, such a translation will be more or less a paraphrase, with one or
another degree of 'modernization', if necessary, to make it easily readable.
Ideally, this kind of translation should be by someone who is an artist in his
own right. If the effect aimed at is something like the original effect, it is
possible for a re-creative translation to be useful, even for the scholar, in
evoking the elusive flavour of the whole work. But it is necessarily impres-
sionistic in two senses. It gives a calculated impression to a new audience,
and it renders only those nuances of the original that the translator wishes
the reader to see relevance in. In any case, for scholarly purposes, a re-
creative translation can never substitute for the original.

The translator's purpose may be to explain and interpret a work — using
the occasion of rendering it into a second language for much the same task

68 INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION

that would be served by a commentary in the original language. Such an
explanatory translation, if cautiously done, will stay fairly close to the text,
but it almost necessarily changes the mood of the work, since words in the
second language grow out of different conceptualizations from the words of
the original language, and the translated text itself, no matter how many
notes accompany it, tends to sway the understanding of the reader. Most of
the current 'serious' translations of belles-lettrist (poetry and literary prose)
works seem to fall into the explanatory category of translation, but with
nuances of the re-creative. Thus, though an explanatory translation may be
exceedingly useful to the scholar, it still cannot dependably replace the
original.

The translator's purpose may be, finally, to reproduce the information
carried by the original work, for the purposes of special study by those who
cannot read the original language. Such a translation attempts to provide an
equivalent communication of the original which readers can then interpret
for themselves. For study purposes, the translation has to be maximally
precise. Special study through translation is legitimate so long as few even
among scholars can master the ever-increasing number of languages in which
significant work has been or is being done. Precise study translations are
generally most appropriate for scriptures, scientific theses, technical philo-
sophy, chronicles, and secondary scholarly essays, though this sort of trans-
lation in fact has its place in many other kinds of material. With ingenuity
and, of course, a complete control of what is routine grammatical pattern or
idiom and what is personal choice in the original writer, precision can often
be combined with great elegance of translation, as has been shown in Islamics
studies by H. A. R. Gibb.

For almost all scholarly purposes, re-creative translation is out of place.
On the other hand, all scholars have to admit their reliance on translations.
In the first place, no scholar can command all the languages now necessary
for him to read in, no matter how specialized his own field. But even more
important, no one person can grasp all the implications of a work — especially
not of a masterpiece — which may be of importance to another reader.
The translator must find an equivalent for every personal turn of phrase
of the original, however superfluous it may seem, and must leave am-
biguities, so far as possible, ambiguous. Such a translation almost neces-
sarily requires a certain number of explicitly technical terms and a few
footnotes or square brackets to pinpoint untranslatable implications. (Need
it be added that a precise translation in this sense cannot be an overly literal
one? Sometimes the most exact rendering of a Persian word or even a phrase
may be, in English, a comma or a semicolon.) The crucial test of the success
of a precise study translation is that, although the translation is free of the
syntax of the original language as such, yet the most natural retranslation of
it into the original language will give back the original form, without pre-
cisions or omissions.

INTRODUCTION TO THE STUDY OF ISLAMIC CIVILIZATION 69

Unfortunately, our supply of re-creative and explanatory translations is
larger than our supply of good precise translations for study. This springs
partly from an inclination of some persons to downgrade the importance of
translating, and partly from a natural desire of translators to be creative in
their own right. But even from the scholar's viewpoint, let alone the lay
reader's, the lack of usable translations is a major handicap to serious work
and proper understanding.

GENERAL PROLOGUE

The Islamic Vision in Religion
and in Civilization

Muslims are assured in the Qur'an, 'You have become the best community
ever raised up for mankind, enjoining the right and forbidding the wrong, and
having faith in God' (in, no). Earnest men have taken this prophecy seriously
to the point of trying to mould the history of the whole world in accordance
with it. Soon after the founding of the faith, Muslims succeeded in building a
new form of society, which in time carried with it its own distinctive institu-
tions, its art and literature, its science and scholarship, its political and social
forms, as well as its cult and creed, all bearing an unmistakable Islamic im-
press. In the course of centuries, this new society spread over widely diverse
climes, throughout most of the Old World. It came closer than any had ever
come to uniting all mankind under its ideals.

Yet the 'civilization of Islam' as it has existed is far from being a clear
expression of the Islamic faith. From the first, pious Muslims themselves
differed as to what the 'best community' should be like. The Islamic vision of
what mankind might be has been seen and interpreted variously: no one ideal
has ever fully prevailed among the Muslims. Moreover their efforts, such as
they were, to build a good society often produced actual results strikingly
different from what anyone had anticipated. Some of the greatest triumphs of
culture under an Islamic aegis have been such as many devoted Muslims could
not look on with favour; and while Islam has seen some outstanding successes
which all could acclaim, it has seen failures at least as outstanding. Those who
have undertaken to rebuild life in Islamic terms have ventured on an enter-
prise with a high potential reward — that of winning through to the best that
is open to mankind; but with correspondingly great risks of error and failure.

Muslims have yet to implement the Qur'anic prophecy fully in all its im-
plications. But they have perennially renewed their hopes and efforts to live
the godly life not only as individuals but as a community. In every age, pious
Muslims have reasserted their faith, in the light of new circumstances that
have arisen out of the failures and also of the successes of the past. The vision
has never vanished, the venture has never been abandoned; these hopes and
efforts are still vitally alive in the modern world. The history of Islam as a
faith, and of the culture of which it has formed the core, derives its unity and
its unique significance from that vision and that venture.

It may be doubted how relevant to historical reality such ideals can be.
Can in fact a world society be effectively built on allegiance to a vision of the

7i

72 GENERAL PROLOGUE

divine? Or must society evolve almost regardless of the ideals or concerns of
any individuals — nay, create and destroy those ideals according to the play of
more casual interests? Certainly, as we shall see, the conscious planning of
idealists has played no great direct role in the social evolution of the Muslim
peoples. Yet the presence of the Islamic ideals (whatever their cause) has
made the crucial difference between the existence of a society that can be
called 'Islamic', on the one hand, and — what might presumably as readily
have been — the perpetuation of earlier traditions in new forms, experiencing
no doubt much of what came to pass under the sign of Islam but without the
peculiar genius which sets off the 'civilization of Islam' as an object of our
interest, our admiration, or our fear.

THE IMPLICATIONS OF THE ACT OF ISLAM

The vision quite naturally, as it is taken seriously in a human mind, unfolds
and materializes itself into a whole complex of life patterns, a whole culture.
Primitively, the term 'islam' refers to the inner spiritual posture of an in-
dividual person of good will. The word islam in Arabic means the act of sub-
mitting to God. (The word muslim, 'submitter', is a participle from the same
verb.) That is, it means accepting a personal responsibility for standards of
action held to have transcendent authority. In this sense, many persons who
have admittedly had no part in the historical community of Muhammad —
among them, all the Hebrew prophets and Jesus and his first disciples — have
been regarded by Muslims as having accepted islam. In any religious tradition,
it is an inward stance in individuals (varying intimately, of course, from in-
dividual to individual) that lies at the heart of all the ritual and myth. It is
this elementary islam, a personal acceptance of godly ideals, which stands at
the heart of Islamic religion, and from which it receives its name.

But the term 'Islam' (capitalized) has come to refer also more generally to
the whole social pattern of cult and creed which, at least for the pious, follows
from or even grows out of the personal islam of the individual devotee; that is,
to the 'religion' in the historical sense. The various elements of the historical
religion can be seen as depending more or less directly on the act of islam of
the individual believer. Some seem to follow almost by logical necessity; and
while others seem to be connected with it only indirectly or even by arbitrary
association, apart from it they would be felt to be meaningless. From the most
essential religious avowals through the realms of uniform cult and law to the
most localized and incidental custom, all can be derived in this perspective
from islam, and all can be included in Islamic beliefs and ways. In particular,
the historical Islamic beliefs about the due place of human beings in the cos-
mos and what they should do there can be seen as stemming logically from the
act of islam; then all else can be derived from these.

Islamic belief has a reputation for being easily understood and while under
careful study it turns out to have its due share of complexity, in fact its most

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 73

essential elements can be set forth rather simply. It is the very point of islam
to own the supremacy of one single God, Who is identified as the God of
Abraham and of Moses and of Jesus, and Whose name in Arabic is Allah. 1 To
make this supremacy meaningful, faith must specify. God is the Creator of the
world and of mankind in it; and He will finally bring the world to an end and,
reviving at the Last Judgment all men and women that have ever lived, will
punish them in Hell or reward them in Paradise according to whether they
have obeyed His will in their lifetimes. To give God's supremacy concrete con-
tent, faith must specify further. God has sent prophets, such as Moses and
Jesus, to summon various peoples to obedience, which consists in worshipping
Him alone and in dealing justly with one's fellow men. The last and greatest of
these prophets, whose message is to all the world and supersedes that of any
previous prophet, was Muhammad of Arabia (d. 632 ce), whose precepts and
example all men and women are henceforth bound to follow, individually and
collectively.

Insofar as much of the morally most decisive human activity is group ac-
tivity, Muslims must work together as a body. But to work together the faith-
ful must recognize one another. True inward islam cannot be perceived from
without. Hence for social purposes one must judge by its most elementary
consequences. Whoever declares that 'there is no divinity but God, and
Muhammad is the messenger of God' (the shahddah) has usually been reckoned
a Muslim and a member of the Islamic Ummah, the community of the faith-
ful, with appropriate rights and responsibilities in common undertakings.
Thus from a fundamental private attitude of soul is derived a concrete social
body and a precise formula of belief to define membership in it. 2

All this must be worked out in massive detail if people are to move from the
general ideal to its actual implementation in the midst of the innumerable
complexities of living. In the course of doing so, Muslims move still further
from the inward core of personal Islam to a vast body of social conventions,
perhaps accepted by most believers rather out of hereditary loyalty than from
any fresh personal idealism.

Above all, the commands of God must be known in detail. Muhammad's
message is embodied in permanent written form. The Qur'an ('Koran' is a
different spelling of the same word) is the collection of the revelations, word
for word, which God made to Muhammad; it has been accepted as literally the
words of God, and every Muslim must know some of it by heart. Hadiths
are reports of the sayings and doings of Muhammad (primarily) , as related by

1 Normally, Allah is to be translated as 'God', just as is (for instance) the Greek ho
Theds in appropriate contexts. Allah is, in fact, the Arabic equivalent of ho Theds not
only for Muslims but also for Christian Arabs, just as 'God' is the English equivalent. To
use Allah in English in referring to the object of Muslims' worship (and not the Arab
Christians') is to imply that Muslims and Christians (or Jews) do not have the same object
of worship; a position not to be adopted lightly. See the note on Allah in the section on
usage in Islamic history in the Introduction.

2 For the pronunciation of unfamiliar terms and names, see the Introduction.

74 GENERAL PROLOGUE

his associates (or perhaps by his representatives); if authenticated, they have
commonly possessed almost equal authority with the Qur'an itself. The
Qur'an and certain collections of hadith reports together have formed the
Muslim scriptures; on their basis Muslim religious scholars, called ulamd,
have set forth in detail what obedience to God means in daily life. (The 'ulama'
are not, properly speaking, priests, but teachers; for every Muslim is authorized
to perform all rites, not just the 'ulama.'.) Their rules of correct practice came
to be called the Shan' ah; in principle it covers even' possible human contin-
gency, social and individual, from birth to death.

Central among the commands of God are those which provide for the cult —
the outward expression in symbolic acts of the individual's devotion. A socially
established faith does not spring simply from a private personal exaltation,
but consecrates generally imposed social obligations; appropriately, then, de-
votion, if not to the ideals, at least to their practical consequences, is expressed
largely in group activity: notably in the ritual worship at the central cult
building, the mosque. Much of the cult, however, even in its group expression,
is carried on outside the mosque; much of it, though socially sanctioned, is
left to the family or the individual — for instance, the requirement of circum-
cision. At this level the cult merges insensibly, by way of private etiquette,
into every sort of family custom, from styles of personal grooming and cooking
recipes to casual superstition. Almost any custom which he recognizes as valid
is likely to be associated by an unlearned Muslim at least vaguely with his
faith — as happens in the case of other religions also. Thus a modern Indian
Muslim may exchange his European-cut street clothes for a more old-fashioned
costume before going to the mosque, feeling the latter costume to be more
'Islamic'; though such a change is not prescribed by Islamic law, and the
older costume itself is simply representative of the part of India he lives in and
is not found in other Muslim lands.

In its narrowest sense, 'religion' can be identified with the cult itself and
with the credal beliefs which are required to give the cult meaning; for it is
only at this level that religion as a social institution usually reflects unmis-
takably a concern with the central impulse of submission to God, without
admixture with other sorts of social interests. Yet religion, in a more meaning-
ful sense, goes beyond cult into every realm, most particularly all moral life,
to just that degree to which its initial impulse is taken really seriously.

Accordingly, the cult requirements are but the most obvious of a wider
range of social practices associated more or less closely with Islamic religion,
sometimes set forth in a uniform Shari'ah law but often varying greatly from
place to place. Most directly tied to Qur'an and hadith reports is a general
system of ethics, which has included a wide body of law of personal status, of
criminal sanctions, and of commerce — never applied exclusively or in its en-
tirety, but always present to the tender conscience. The sense of membership
in a world-wide community with a historical mission has been very strong,
and the obligation of Muslims to help one another has been an explicit part of

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 75

the faith, particularly in the case of jihad, a war against non-Muslims sanc-
tioned in certain circumstances by religion. Almost as closely tied in with
religion have been certain social customs, such as the practice (never universal,
and now perhaps in course of disappearing) of keeping women in strict social
segregation from men. Less closely tied to religion is the use of the Arabic
script for writing whatever local language the Muslim uses — a practice, almost
universal in pre-Modern times, which helped to set off a series of 'Islamic'
cultural languages from sometimes almost identical 'non-Islamic' ones. Still
further from a religious prescription is the development of certain typically
'Islamic' decorative patterns of a complex linear type, especially the 'arab-
esque', often involving ornamental use of the Arabic script, whether in mosques
or in buildings unrelated to the cult. With such features we are gradually
leaving the realm of what are clearly points of Islamic religion as such and
moving into the wider realm of culture at large.

Associated with Islam by extension comes to be a whole vocabulary of art
motifs, not merely the arabesque but even related figural styles; a customary
expectation of certain social and political forms and standards; above all, a
classical body of literature ranging, by slight gradations, from devotional
and legal works through works of theological disputation to works of pure
metaphysics and natural science; from the history of the Prophet and his
community through works of moral edification and general information to
every sort of belles-lettres in prose and verse. All were written under the sign
of Islam and, to a remarkable extent, all have managed to justify themselves
as in some way contributing to the fully Islamic life; they have been carried
wherever Muslims have gone and, transmitted from generation to genera-
tion, have formed the common background of literary culture shared among
all Muslims of cultivation, those who maintained the norms of Islamic society.
Not only what may be called the religion proper, then, but the whole social
and cultural complex associated with it — indeed, at the most extreme
extension, the totality of all the lifeways accepted among any Muslims
anywhere — may be looked on as Islam and seen as a self-contained whole, a
total context within which daily life has proceeded in all its ramifications.
All can, in some sense, be derived as consequent upon the initial posture of
islam, of personal submission to God.

As the accompanying chart of the twentieth-century distribution of
Muslims shows, though Islam now counts only about a seventh of the world's
population, it is unique among the religious traditions for the diversity of the
peoples that have embraced it. It began among Arabs in a hot and arid
climate, but very early became international and has since taken root in the
coldest north and in the wettest tropics. Yet everywhere it went there has
been a continuous pressure toward persuading all Muslims to adopt like
standards, like ways of living, based on the Islamic ideals prevailing at a
given time. There has never been any central world organization of Muslims
after the first generations. But even now everywhere Muslims are noted for

a

i— i

I

o

■-CJ
H

a
8

§

o

On

(SI *-

VI

Vl £

01 B

cd en

O J .O

co

co O

■« 8

o a,

g P-c

o

M s
£3

■g 3 £
■a n o u

U CTJ «*

bo u • - ni

co co a> ^j -,

la

3 03 C cd .K

(3 5

Bfip

5".

"" "3
u o h

5 S fc

-a
a

18 ™

fi.Si

•■g-s

6 °

sS
IB

C o

■° s

03 •£*

Ot-,

o P

2 "M

o o

gH»>

..- « s

S -b
1*

^.■2

W a*
-is o

in fl)j3

13

B

cd

3

be

O

13

-P

P 3

<

75

25^

bo co

cd cd

S o

co a

as
s-s.1

la

is

•a g-S
S3 g 3

i>, CD +>
TJ S *

•3 B

^^ 2

a »

S N -

■s<! a

.fcp'g ra

■« §■§

" ^ Co
bo

Iq

u

bo

M

CO

3"

0)

o
a

g

a

'-P

a

ca

O

>

o

CO

<0 iT

QJ

Cli S

^55 g

.2 >»

CO S'C

o .>"3
m a no

•3.2, g'bo
3 co N 3

O ccs

£

O

s
■a

CO

CD

D

B

S

a>
o

CO

J2

3

<E>

X

u

a

be

S3
1

Rep-
erita;
influ

CD
CO

.g

J3

CO

^

a

T3
B
CO

B

CO

B

d

+j

CO

CO

en

T)

jjfl

El

,3

B

'o

fl

S

CO

>

O

>»

C

bo

CO

□JO

B

CO

" o-"

2-i?

■S CO
to 0) co

sHPQ
10

is
s

bl

J3 N
Si- 1

b" b"

CO K

-,13

bo

B

eg . .

a a

Hi

^ .§ ^f 2 ef

bo

■SM

03
Si

d o

d , ,

03

ol

» P.

3

o

iS fcj d

W.£to

■a

s

Is

O 3

d

T)

4->

CI

3

d

03
3

03

g

'-+3

a
o

'3

13

o

'5

B

a

o

•a

M

•3

h

,3

o

d
>

>

d
si

Is

»> 5?

,3

3 w-g "3
aatis
Pl, u 3 «

— - en

Q T3 d
*5-i 3-3

0)

1-1 .2

I If!

o « ° o

|| "..si

nJ q ,±fl .,., ,_,

u-frc * °
s -a .§£-+*

j3 o -D ■" -p
5 C » C 3
r! "e b 2
"^•^
!>. a .. n en
+J „j en S5 a

O .2 "3 m -3
.3 "3

V, *

oi> B

I -8 S3

o E

■saS

o d

.Jo.

>;c.a
'S ^S

O -rH T3

3 +j g
11^

8«.

o 2

CO 3

03 d ,,

boo d

3 ° °

^ -

a> tub rf

CO d

o P.

w d

J!

o

o
H

^3
•O

•a

3

w

ho

a
o

B

id

Sp^ffi2S OB§u S

Jf.S'&S gfi

82

in g

-a

a

d

3 S

SM
.B o,

Odd

2rs ft =3

3
d

g B
co en

42 bo
^ a

d o

a 3

cj «j C

*-■ n3 nJ

w

t3 *n

3'C-—

30§

in fi «

a^ B

■a

3
d

• S'O

a » S

a » »„

<o

03 te

d ^ .

S g u
« B"> <d-
fe<ja &

f «>3 °

d "

^B

■* o
•o <->

o, «

f *

CO O

ST3
§ d

« E

S CJ

o o
a n

^ S-o
*e o n

^

■3a

aj d

O 03

s

8

& > b

O > 3
H ,4->

■ h rt S3
03 R o
03 i-l

d ,3,3
« > o

.a*. "

p d 03

o ,3 ^a

J* 3 +■

C o

u a, s

. 03 03

■a > r°

d bog

Soo
• S ^ B

&•«■§

S °«

^ T3X-.
03 03 o
H -P *

d o m

03 03

a ■£?>»

g P d

03 ^ «
d "M

*-" m b
3 O O

j8 GENERAL PROLOGUE

their keen consciousness of the world Muslim community; they are moved
by a sense of universal Muslim solidarity, and maintain in the most diverse
geography not only the essential distinctive Islamic rites — including the
great common pilgrimage to Mecca where all nations may meet — but also,
to some degree, a sense of a common cultural heritage.

At' any given time, in any pious mind, Islam could thus seem a timelessly
integral ideal whole. Islam as such could not change, though its practice
might here and there be more or less corrupted. Such a viewpoint is not purely
a subjective illusion. Insofar as Islam, as a religious teaching, can be legiti-
mately formulated as an integrated overall doctrine based on crucial beliefs about
the nature of existence (in particular, on a belief about the Creator implied in
the basic act of islam), then to that extent it imposes its own limits on its own
future. Such a doctrine, though it may undergo some further development or,
more likely, corruption cannot be too radically altered in substance or in
spirit without the original teaching itself being denied, and therefore a new
ground for faith being implicitly required.

Within a given social context, at least, this integral view may well hold.
The Shari'ah law, for instance, in the form it has taken, as a comprehensive
corpus of life regulations, nowadays seems tied to certain outdated social
assumptions. Many Muslims would like to 'Modernize' it. But it has yet to be
shown how far it will be possible for modern Muslims to modify the Shari'ah
in its more fundamental presuppositions without, in effect, abandoning any
serious allegiance to the traditional hadith reports on which it is based. Yet
the hadith reports have served to interpret the Qur'an ever since the Qur'an
ceased, at Muhammad's death, to be its own continuing interpreter; it is
questionable, then, how far the hadith reports can, at this date in history, be
separated off from the Qur'an as dispensable, without decisively abandoning
hope of grounding life on the Qur'an and in the islam it calls for.

In a less precisely formulable way, the same is true of the culture generally,
of the overall complex of lifeways associated with the religion. In any culture
can be seen a distinctive manner of fitting life together, which gives it a dis-
tinctive tone or style. New ways introduced either from within or from out-
side may be assimilated to the cultural style and fit in easily; or they may
prove incompatible with it and arouse endless difficulties, occasioning a
round of readjustments throughout the culture or even disrupting it alto-
gether (as has happened with some non-literate peoples which disintegrated
when their members met modern lifeways they could not learn to cope with) .
An integral conception of Islam as a total culture highlights its distinctive
style, its cultural integrality as an indissolubly coherent whole, by tracing all
its ramifications to what seem to be indispensable foundations. One is fore-
warned of points of incongruity and strain in the cultural fabric. On occasion,
what comes to be lived as Islam, in particular cases, does violate the integrality
of Islamic life: that is, it proves inconsistent with more fundamental cultural
presuppositions of Islam, as Islam has been developed; and it is bound to

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 79

produce a conflict, therefore, which will require some sort of psychological or
historical resolution. In some sense, Muslim life has had an integral character
as Islamic, not to be violated with impunity.

THE DIALECTIC OF A CULTURAL TRADITION

What has been felt as Islam, however, considered historically, in all its ramifi-
cations and even in its most central implications, has of course varied enor-
mously. The very comprehensiveness of the vision of islam as it is unfolded
has insured that it can never be quite the same from one place or one time to
another. 3 Empirically, any particular formulation of what the fundamental
consequences of the act of islam must indispensably include, would find serious
Muslims to reject it — as would a corresponding attempt with regard to any
other religious tradition. Still more would this be the case for the culture
generally. For historically, Islam and its associated lifeways form a cultural
tradition, or a complex of cultural traditions; and a cultural tradition by its
nature grows and changes; the more so, the broader its scope.

Tradition can cease to be living, can degenerate to mere transmission. A
recipe for a holiday pastry may be 'traditional' in the sense merely that it is
transmitted unaltered from mother to daughter for untold generations. If it is
merely transmissive, a sheer habit, then any change of circumstances may
lead to its abandonment, at least once the mother is gone. But if it is vital,
meeting a real need, then the tradition will be readjusted or grow as required
by circumstances. A living cultural tradition, in fact, is always in course of
development. Even if a pattern of activity remains formally identical in a
changed context, its meaning can take on new implications; it can be gradually,
even imperceptibly, reconceived. A pastry first made when all foods were pre-
pared at home inevitably becomes something very different when it alone is
home-made, though exactly the recipe be used. To cling to the recipe then
requires, or perhaps produces, a new point of view toward the pastry. But
even without so drastic a change in circumstances, the recipe and its use will
prove to have a history. Even in primitive life, over the millennia or even
only the centuries, fuel differed, or water, or the quality of the utensils. Even-
tually, if the tradition was genuinely alive, some cook found that the recipe
itself could be improved on in the changed conditions. As she did so, she was

3 Wilfred C. Smith, in The Meaning and End of Religion (New English Library, 1966),
has pointed out that the very notion of 'a religion', as an integral system of belief and
practice held to be either true or false, is relatively recent as compared with the notion
of 'religion' as an aspect of any one person's life, which may be more or less true as that
person is more or less sincere or successful. Even in the Irano-Semitic sphere, where
'religions' were earliest and most sharply set off as self-contained total communities,
the notion of 'a religion' as a system was slow to prevail and has become dominant only
in quite modern times. He suggests that what we generally have to deal with are
cumulative traditions through which religious faith has been expressed. I am indebted
to him for sharpening my awareness here.

80 GENERAL PROLOGUE

not abandoning the tradition but rather keeping it alive by letting it grow
and develop.

Living societies seem never to have been actually static. With the advent of
citied and lettered life, this dynamic aspect of cultural tradition was intensi-
fied; or, rather, the living tradition-process was speeded up and became more
visible, so that generation by generation within each tradition there was a
conscious individual cultural initiative in response to the ever-new needs or
opportunities of the time. It is in what is called 'high culture', in relatively
widely shared cultural forms at the literate, urban level, that social tradition
has unmistakably shown itself as a process of change. Yet even in 'folk cul-
ture', the culture of peasants or even of non-lettered peoples, cultural tradi-
tions share substantially the same dynamic force which is more visible in high
culture. 4

In general, then, but especially in the high culture of pre-Modern citied
societies, which has been the primary milieu of Islam, we may describe the
process of cultural tradition as a movement composed of three moments: a
creative action, group commitment thereto, and cumulative interaction with-
in the group. A tradition originates in a creative action, an occasion of inven-
tive or revelatory, even charismatic, encounter: for instance, the discovery of
a new aesthetic value; the launching of a new technique of craftsmanship; a
rise to a new level of social expectation, one man of another; the assertion of a
new ruling stock or even the working out of new patterns of governing; or,
in the case of religion, an occasion of fresh awareness of something ultimate in
the relation between ourselves and the cosmos — that is, an occasion of spiri-
tual revelation, bringing a new vision. In accepting the Qur'an and its chal-
lenge, Muhammad and his followers opened themselves to vast new considera-
tions of what life might mean, which relegated their former concerns to frivo-
lity; their act of acceptance was thus intensely creative.

Such occasions are creative partly through the quality of the objective
event itself, in which there must be something which genuinely answers to
universally latent human potentialities. The human impact of the Qur'an as a
sheer piece of writing is undeniable. At the same time, the occasions are crea-
tive equally through the particular receptivity of their public, of those who
take up the creative event and what it has produced and assign it value,
finding in it something which answers to their particular needs or interests,
material or imaginative, so that it becomes normative for them. The Qur'an
spoke not only in the language of but to the personal and social needs of a
particular group of Arabians, of Meccans and Medinese, with particular social
and moral problems. By their responses, positive and negative, they built con-
crete meaning into what might otherwise have remained on the verbal level as

* Throughout this chapter, I have tried to take advantage of the increasing sophistica-
tion of anthropologists about cultural processes; but since I am dealing, as a historian,
primarily not with folk culture but with high culture, I have had to develop my own
theoretical framework here.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 8l

general exhortations and observations. Without such response, which indeed
is presupposed in the later portions of the Qur'an itself, it could at most have
become a striking but otherwise inconsequential piece of literature. This
double aspect of the creative action is nowhere more crucial than in the reli-
gious life, where revelatory possibilities are doubtless latent in many an event
that passes almost unnoticed; but revelatory experience occurs only when
enough persons are ready to receive the impact of a given event and allow it to
open their eyes.

The second moment of a cultural tradition is group commitment arising out
of the creative action: the immediate public of the event is in some way
institutionalized and perpetuated; that is, the creative action becomes a point
of departure for a continuing body of people who share a common awareness
of its importance and must take it into account in whatever they do next,
whether in pursuance of its implications or in rebellion against them. Such
was long the case of Occidental artists vis-a-vis Italian Renaissance painting,
for instance. In a tradition of liberal education built around an agreed-on core
of classics, the commitment becomes even more binding, still more so in a
tradition of law. At its most effective the commitment becomes an allegiance.
Thus Islam could be defined as commitment to the venture to which Muham-
mad's vision was leading; which meant, concretely, allegiance to Muhammad
and his Book and then to the continuing community of Muhammad, or at
least (later) to a supposed faithful remnant of that community. The allegi-
ance came to be marked by such symbolic gestures as utterance of the
shahadah, the formula of Muslim belief, or performing worship toward the
qiblah, that is, toward Mecca; which acts have been sufficient to establish a
person as committed to the social and juridical consequences of being a Mus-
lim, whatever the extent of his inward islam.

This group commitment retains its vitality through cumulative interaction
amongthose sharing the commitment; above all, through debate and dialogue,
as people work out the implications and potentialities latent in the creative
event to which they are bound. In the arts, the solution of one problem — it-
self arising from within the artistic tradition — is witnessed by other artists,
who may adopt it or respond to it with alternative solutions; and these solu-
tions open up new opportunities and new artistic problems in their turn. In
philosophy and in science, the transmission of what has been done and found,
especially in the case of the great initiators like Plato or Ptolemy, is but the
preliminary essential to the continuing cumulative dialogue, the response and
counter-response, which is the purpose of such transmission and without
which the transmission itself will gradually cease. The like is true of the inven-
tive and elaborative development of an economic order; and above all of
political life, built of the thrust and parry of contrasting interests, each party
striving to turn the accumulated heritage of a major state formation to its
own advantage, and in doing so shifting the pattern of the heritage which the
next special interest will work through.

82 GENERAL PROLOGUE

The implications of an initial event may be relatively particular and func-
tional — may have reference to the development of historical circumstances as
the community faces them. Once the early Muslims had conquered the region
from the Mediterranean through Iran, a common corpus of law and custom
was required if the community was to hold together and maintain its position:
some sort of Shari'ah law was needed. And if this was to be worked out in har-
mony with the experience that had brought the Muslims together and assured
them of their highest ideals, building a Shari'ah could require, in turn, the dis-
covery of hadith reports, of reports about the earliest Muslims, which could
relate the Qur'anic challenge to the practical needs of the subsequent empire.
Gradually, over centuries, pious Muslims building one on another's work
found such hadiths and created such a Shari'ah. Much of their work meant
coming to terms at least as much with their own society as with the Qur'anic
challenge.

Or else the implications of the creative event may be relatively universal
and logical — may stem, that is, from the very nature of the event, from its
inherent potential for enlarging the resources of any human beings seeking
truth in their lives. Thus whatever may have led a man to commit himself
initially to Muhammad's community (and it was likely to have been much less
than a total apprehension of Muhammad's own vision), this public commit-
ment necessarily led him further in the direction of that islam which is the
private affair of the soul; and of seeking an articulate conception of prophecy
and of God, such as would make intelligible the islam he found himself being
led towards. It was in dialogue, in an exchange of insights and of objections
with others in a like situation, that he came to realize what sense of himself
and of the universe would serve and what would not. In such searching, he
might be confronting something in Islam, and perhaps something in the
human spiritual condition, more or less relevant under any circumstances.

Relatively circumstantial and functional or relatively inherent and logical,
in either case the implications of the initial commitment could be worked out
fully only as the initially creative vision was confronted ever afresh from a
new perspective in the course of cumulative interaction and dialogue among
those to whom the initial events were meaningful.

But not only does a developing interaction arise out of an initial point of
creativity; that interaction, that dialogue, itself is made up of a sequence of
creative actions and of commitments stemming from them — secondary ac-
tions, secondary commitments, up to a point, but genuine actions, encounters
and discoveries, all the same. Rather than being an ideally fixed pattern
which might almost be deduced from the initial creative event, supplying a
determinant body of ideas and practices, a cultural heritage forms a relatively
passive setting for action. Within that setting, any given juncture may bring
a fresh turn of orientation ; or at least its outcome will be relatively unpredict-
able, for the same setting will allow for varying sorts of actions according as
circumstances, temperaments, and problems vary.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 83

Thus, within the dialogue launched by the advent of Islam, almost from the
start there came to be conflicting sets of presuppositions about what Islam
should involve, each producing its own commitment and its own dialogue.
Around the charismatic figure of 'Alt, companion of Muhammad and the
fighting man's hero, the common soldier's sense of Islamic justice crystallized,
even in the first generation, against what were felt to be the backslidings of a
wealthy clique which had got control of the Muslims' conquests for its own
benefit. This sentiment, sealed in the blood of rebellion, issued in a deep
loyalty to 'All and later to his house. The resulting movement was called
Shi' ism. The bulk of the Muslims were felt to have gone astray in rejecting
'All's leadership, only a faithful remnant holding to the original vision. Genera-
tion by generation, the widest possible implications of such a loyalty in such a
remnant were worked out among the Shi'i Muslims, implications for social
justice not merely in the soldiers' cause but in all fields, and (since life is whole)
implications also for the personal devotional life, for metaphysics, and for the
whole range of Islamic concerns. At the same time, the stirring and demanding
experience of forging an effective unity among the Muslim community, despite
such partisan pressures as that of the Shi'is, was yielding a contrasting loyalty,
called Sunnism: a loyalty to the Muslim community as it had come histori-
cally to be constituted, despite all its faults. This loyalty likewise had its per-
vasive implications, gradually worked out in a long dialogue among the
Sunni Muslims.

Thus arose within the Islamic setting two differing sub-traditions, sub-set-
tings for dialogue, within which the implications of the original advent of
Islam were being worked out in contrasting ways. And each secondary
dialogue, like Islam itself, began with a point of creativity: within each, it
was duly noted that there was something divinely guided or even revelatory
— though of course not in the same degree as in the Qur'an itself — in the
position of 'All, the hero, or in that of the Ummah, the general community,
respectively. As may be seen from the map of the distribution of Shi'is, the
two contrasting traditions have endured to the present. (In our day more
than a quarter, perhaps a third, of the Muslims in the central lands of 'old
Islam' from Nile to Oxus are Shi'is. The rest there — as well as most Muslims
in the more outlying lands — are chiefly Sunnis, who therefore form nine-
tenths of the Muslims in the world at large.)

THE DIVERSITY OF ISLAM IN THE NEXUS OF GENERAL HISTORY

As the Islamic tradition developed within its own terms, it was likewise inter-
acting with other cultural traditions which were already present among the
populations in which Islam was adopted. No tradition is isolable from others
present in the same social context. (A culture, indeed, may be defined as a
complex of interdependent traditions.) A creative event in any sphere of life
is likely to have consequences in many fields; in any case, the interaction and

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 85

dialogue in religious or artistic or political or scholarly settings overlap and
merge because the important problems that arise are rarely so technical that
only one sort of tradition proves relevant. The sub-traditions into which
Islam seemed to divide, even when mutually hostile as were those of the
Shi'is and Sunnis, were not in fact exclusive; it was common for an in-
dividual Muslim, almost from the start, to share to some degree in both of
these. Still less did people cease to take part, when they became Muslims, in
the other ongoing traditions which moulded their lives, except in the few
cases where the commitment involved was explicitly incompatible with the
Muslim commitment: as in some, but not all, aspects of other religious tradi-
tions. For the rest, Islam found itself in a vital and multiple cultural environ-
ment. It was only as it entered into these other dialogues, in fact, that it
could become significant for cultural life at large.

The artistic traditions of which Muslims found themselves at first patrons
and later also practitioners had been launched in pre-Islamic times. Only as
Muslims (rather gradually) found special viewpoints toward elements in these
traditions — viewpoints suggested as readily by their particular social situa-
tion as by any more strictly Islamic inspiration — did the continuing develop-
ment of those traditions introduce any features that might be called Islamic.
The same is true of scientific and philosophical traditions, of traditions of com-
merce and of crafts and of public administration, even of pious legend and of
ascetic practice. Always the presence of Islam made itself felt only gradually
as Muslims discovered points at which it would be relevant within the settings
provided by the various other traditions.

Such a difference as that, often noted, between Islam in a majority-Muslim
environment (such as the Arab lands) and Islam in a minority-Muslim en-
vironment (such as India) is not, then, a matter of the degree to which Mus-
lims are Muslims. It cannot be reduced to the difference between stale custom
and enforced alertness, as an Indian Muslim might suggest; or between inbred
mastery and latter-day imitation, as some Arab Muslims might feel. It will
also be the difference between two equally genuine responses to the overall
spiritual challenge carried in the Islamic dialogue in the midst of two different
cultural environments. Indian Muslims, for instance, would read at least some
books by other Indian Muslims, being stimulated by the special Indian pro-
blems those writers were contending with; and would respond positively or
negatively to the creative experiences to which the writers bore witness. Arab
Muslims likewise had their own subordinate dialogue with its own special
Arab problems and encounters.

When we look at Islam historically, then, the integral unity of life it seemed
to display when we looked at it as a working out of the act of islam almost
vanishes. In such ever-renewed dialogues, among settings formed apart from
Islam at all, is not anything possible provided only it possess a certain general
human validity? We can no longer say that Islam eternally teaches a given
thing, or that another thing is necessarily a corruption of Islam. Such judg-

86 GENERAL PROLOGUE

ments a believer may feel himself able to make, but not a historian as such. At
a given time, in a limited milieu, perhaps, Islam may form a relatively de-
limited and inviolable pattern. But over time, and especially on a world scale,
any particular formulation of thought or practice is to be seen as the result of
how the ever-changing setting formed by the Islamic tradition is reflected in
particular circumstances and in relation to all the other cultural traditions
present. 5

Even the persistent pressure which has existed toward some sort of unity
or uniformity among Muslims everywhere is not to be seen as an inevitable
attribute of Islam as such. Rather, it arose out of the relationships which
held, at various times, between the ongoing Islamic tradition and the other
cultural traditions among which it was developing. In the time of the first
Arab conquests, the sense that the Qur'anic vision demanded some sort of
unity among the war bands of the faithful was sufficiently expressed in a
demand for a single supreme authority among them, a caliph who would lead
their worship and their wars. The notion of a caliph was never abandoned, but
later — when most Muslims were merchants or craftsmen or peasants — it be-
came merged in the ancient notion of the disinterested royal arbiter among
the several classes of a settled society; eventually Muslims could admit several
caliphs at once. Muslims by then, however, represented numerous differing
lltural traditions; those who were concerned to fulfill the Qur'anic vision,
ccordingly, found it necessary to demand a new sort of unity: a unity in the
Listoms of social intercourse, on which could be built in common a just society,
till later, under the disruptive impact of Modern times, a third sort of de-
mand for unity seemed required. Tendencies toward what has been called
'pan-Islamic' sentiment have represented a hope that the various peoples of
Islamic faith, having found themselves in a common subjugation to the
Modern West, would co-operate politically in their common interests, how-
ever diverse they remained culturally.

What then is Islam? Can we study it as a meaningful whole? Is it more than
the name for a hope, and a few common symbols? Clearly, yes: but only in the
way that any cultural tradition, whatever its internal contradictions, is a
whole. However diversely it develops, or however rapidly, a tradition does not
lend itself indifferently to every possible opinion or practice. It imposes limits
which are none the less enduringly effective for being impossible to formulate

5 Not every scholar, and certainly not every Muslim, will be happy with so strong a
limitation as T put on the existence of any eternal 'true' Islam. It is conventional, in fact,
for scholars to distinguish more and less 'true' or 'genuine' Islam among the various
forms which Islamic consciousness has taken; they often label some forms 'orthodox' and
use other categories, implicit and explicit, which presuppose such distinctions. There
is a certain validity in such usages, but only, I think, when very carefully circumscribed.
Partly, of course, it is a matter of definitions. I have used 'Islam' in the way which, it
seems to me, most genuinely answers to its prevailing usage in ordinary contexts. Here
there can be more practical accord than on the level of global theory. Perhaps my usage
should always tacitly presuppose some such adjective as 'historical', as against 'ideal'
or 'metaphysical'.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 87

in advance. Dialogue within a group, indeed any dialogue, is scarcely possible
if everything is put in question at once. If a new insight is to come at one
point, it can be clarified only against a background of received insights, held
for the moment as if they were constant, or at most shifted only in per-
spective. The same is true of any sort of interaction: an innovation at one
point is feasible only if it can be assumed that at least some other points will
remain fixed. At any given time a minimal cultural integrality can be
assured at least by those features of the cultural setting which even those
men take for granted who think they are out to change everything. Over the
generations, innovations may indeed appear at any point. Yet so long as
there is a common commitment to an initial point of departure which all
acknowledge, and to the continuing body of persons which shares that
acknowledgement, the dialogue will retain common features, even though
these are not necessarily those most visible at any given time. For to the
extent that the dialogue is cumulative, every later comer having to reckon
both with the point of departure and with the later debate, there will needs
be a common vocabulary of ideas (or of art forms or institutional principles
or whatever) which will include somehow all generations concerned. It is
this integrality of dialogue that can provide an intelligible framework for
historical study.

THE UNITY OF ISLAM AS A RELIGIOUS TRADITION, AND ITS LIMITS

We can distinguish two levels on which there has been effective continuity in
such dialogue among Muslims over the centuries: that of religion and that
of civilization. Throughout this work, we will be dealing with a religious
tradition and with a civilization; we must clarify briefly here what we mean
by religion and what by a civilization; and what sort of relations can exist
between them. Cultural continuity among the Muslims is most visible on the
level of what we call 'religion'. The Islamic religious tradition, for all its diver-
sity, has retained a certain integrality; distinctly more so than, say, Chris-
tianity or Buddhism. But we will find that this religious unity among Muslims
is but one expression of a wider cultural unity. This wider cultural unity is
historically, doubtless, the more fundamental.

The religious unity must be recognized first; and among Muslims it already
carries with it much — but by no means all — of the realms of culture that in
some other contexts might not be considered religious at all. At the very least,
all Muslims must come to terms with the Qur'an; and in doing so they must
not only talk in part a common language, but must find themselves faced with
at least partly common challenges. Hence the whole range of what appears
under the Islamic name can be relevant to seeing the full implications, positive
and negative, of a personal commitment to its founding events and their latent
challenges. But because even the most diverse Muslims have sometimes read
not only the Qur'an but each other's subsequent books, the tradition has had,

00 GENERAL PROLOGUE

in fact, more integrality than this minimum. Consequently, any serious ques-
tions about that commitment, its implications, and even its historical develop-
ment will prove interdependent; the tradition must, then, be studied as a
whole.

As we have seen, this religious commitment and dialogue carried far. It con-
cerned not merely a corner of people's lives, reserved for moments of special
exaltation or of special despair. It reached pervasively into daily living. How
far it reached, we must see. For we cannot arbitrarily set off the sphere of
'religion' in general in advance, even if we give it a broad mandate; we cannot
assume that such-and-such a type of activity must be 'religious' and other
types of activity not. Indeed, so variable is the sphere of 'religion' in different
cultures that a common term for all that we call 'religious' is really justifiable
only by invoking a series of extended meanings.

In a person's life, we can call 'religious' in the most restricted sense (in the
sense of 'spiritual'), his ultimate cosmic orientation and commitments and the
ways in which he pays attention to them, privately or with others. Properly,
we use the term 'religious' for an ultimate orientation (rather than 'philo-
sophical' or 'ideological'), so far as the orientation is personally committing
and is meaningful in terms of a cosmos, without further precision of what this
may come to. In an Islamic context, this has meant, in effect, a sense of cos-
mic transcendence, and we may apply the word, more concretely, to efforts,
practical or symbolical, to transcend the limits of the natural order of foresee-
able life — that is, efforts based on hope from or struggle toward some sort of
'supernatural' realm. 6 Then we may call 'religious' (extending the term a bit)
those cultural traditions that have focused on such cosmic commitments:
cumulative traditions of personal responses to presumed possibilities of trans-
cending the natural order. Such traditions take off from events people have
found to be revelatory of such possibilities. But in common usage the term is
extended still further. A person's actual 'religious' life does not necessarily
consist in creative cosmic commitment; it consists in his participation in
religious traditions as given to him — -in any aspect of them, whether he per-
sonally cares much for the initial spiritual impulse, or carries on only those
aspects of the tradition that were secondary to and supportive of the primary

6 For some purposes, one can apply the term 'religious' wherever an experience of the
numinous or a notion of the transcendent (commonly linked thereto) becomes life-
orientational. But in different life-orientational traditions, and even within the same
tradition, the numinous-transcendent can play very different roles, or be absent even in
primary moments. Hence, for the purpose of classifying the traditions (if one must), a
more general definition will help: one can apply the term 'religious' to any life-orienta-
tional experience or behaviour in the degree to which it is relatively most focused on the
role of a, person in an environment felt as cosmos. Any developed life-orientational outlook,
however much oriented to culture or history to the exclusion of concern with natural
cosmos, deals with personal life and has some interpretation of the cosmic whole ; but the
term 'religious' would not be applied to the central experience and behaviour of, e.g.,
Marxism, since, in contrast, say, to 'atheistic' Buddhism, the relation person-cosmos
plays a relatively slight role there.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 89

spiritual commitment. For historical purposes, it is not very feasible or even
desirable to separate out these different extended usages.

Accordingly, though what we call the 'religious' traditions have been
centred on such ultimate commitments, they have not — as actual historical
traditions — been reducible to them. Whatever its central concern, Islam has
come to imply much else besides, not necessarily deducible from its concern
with transcendence; and as a historical tradition, Islam must be seen integrally
in its own terms, constituted by all these diverse things. It is not because
Islam embraces a certain cultural sphere but because, in the spheres it does
embrace, it happens to put central an effort toward transcendence, or toward
cosmic meaning, that we call the Islamic tradition 'religious', and participa-
tion in it 'religious behaviour'. Even narrowly conceived as cult and creed,
Islam contains far more than — and also less than — what we might suppose
was necessarily implied in an effort toward transcendence: in, for instance, the
direct personal act of islam. It can be religious behaviour, for instance, to
extol one's own religious heritage — in this case Islam — and to denigrate other
people's; but this need not indicate a very lively sense of personal submission
before the God who created all. Indeed, Islam may be highly cultivated, for
social or intellectual reasons, by persons who have had little such religious
experience of their own. (We must also, of course, be ready to recognize on
their own merits the norms of all the sub -traditions that diversely developed
Islam — all the contrasting positions that were 'Islam' to one or another group
of Muslims.)

The reader will find that Islam, rather more than Christianity, tended to
call forth a total social pattern in the name of religion itself. None of the great
religious traditions of this type has been content with occasional acts of wor-
ship in consecrated buildings — all have hoped to form men and women's daily
attitudes and conduct. In principle, any religious allegiance might make de-
mands on every aspect of life to such a degree that a religious body could
constitute a complete society, its way of life a self-sufficient culture. But Islam
especially has tended to make this kind of total demand on life. In many
spheres, not only public worship but such spheres as civil law, historical
teaching, or social etiquette, Muslims succeeded quite early in establishing
distinctive patterns identifiable with Islam as religion.

But even Islam could not be total. Even in these preferred spheres, speci-
fically Islamic patterns rarely prevailed exclusively; and in many other spheres,
such as trade or poetry, the articulated religion had to be content to lay down
limits which the merchant or poet should not overstep. Otherwise, these as-
pects of culture were cultivated, in substantial autonomy from any particular
religious allegiance. What was religion and, in particular, what was Islam,
was always, if diversely, kept consciously distinct from the total culture of
Muslim society. In even the most pious man's life there was much that he
could not call religious.

The wider cultural life of Muslims, their civilization, had its own historical

90 GENERAL PROLOGUE

integrality, which was not simply an extension of the specifically religious
unity of Muslims. Indeed, this wider cultural complex included the Islamic
religious life as but one facet, albeit a central one. For around the actual
Islamic religious tradition was formed, historically, an overall culture not re-
stricted to what was 'religious', even granting the wide sphere claimed for
Islam as religion. It was not only the Qur'anic challenge and its consequences
that Muslims confronted together, but also a whole series of historical events
and problems in every sphere of life. In Islamic times, to an important degree,
the arts and sciences (for instance) of Muslim Turks and Persians, of Muslims
of Egypt and of India, were interdependent; moreover, they were clearly dis-
tinct from those of lands outside the Muslim sphere even when they defied the
religious convictions of most pious Muslims. Further, since the cultural tradi-
tions which together made up the civilization associated with Muslims often
depended little, directly, on the Islamic tradition as such, they were by no
means restricted to Muslims. Many non-Muslims — Christians, Jews, Hindus,
etc. — must be recognized not only as living socially within the sphere
of the Muslim culture; they must be recognized as integral and contributory
participants in it, engaging actively in many of its cultural dialogues. At the
same time, some groups ardently Muslim in religion (for instance, in China)
were only limitedly influenced culturally by this cultural complex. The scope
of the historical civilization, then, was not only distinct from the religion as to
field of activity; it was not even coextensive with it in time and space, and as
to the populations involved. Here we have, then, a second, more inclusive level
on which cultural traditions carried by Muslims form an interrelated whole to
be studied as such.

This wider level of integrality, however, has been profoundly influenced by
the religious level, especially among Muslims, not because it is an extension of
it but because of the strategic position in which the carriers of the religious
vision have found themselves within the civilization, notably in the high cul-
ture which gives definition to a civilization in the most meaningful sense of
that term.

A CIVILIZATION AS EXPRESSION OF FORMATIVE IDEALS

In this work we shall speak more of masterpieces of art, and dynastic policies,
of religious geniuses, and scientific discoveries, than of everyday life on the
farm and in the kitchen. Hence we will include in our scope those peoples
among whom a few privileged men shared such masterpieces and discoveries,
however much those peoples differed among themselves in farmwork or in
homemaking. This may seem like arbitrary preference for the spectacular. I
believe it answers to a legitimate human need to understand ourselves. In any
case, we must be clear as to what we are doing, and its consequences.

The wider culture associated with Islam has been as highly differentiated
and heterogeneous as any; has been, in fact, the sum of many cultures, or at

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 91

least of aspects of them. The peoples concerned, flourishing from the time of
Muhammad in the seventh century to the present, have extended in space
over half of the Eastern Hemisphere of our globe. They have been corres-
pondingly diverse in language, climate, historical situation, and national cul-
ture patterns. It is such a compound culture that we call 'a civilization': that
is, a relatively extensive grouping of interrelated cultures, insofar as they have
shared in cumulative traditions in the form of high culture on the urban,
literate level; a culture, that is, such as that of historical India or Europe
taken as cultural wholes. Such groups of peoples have varied greatly among
themselves and yet have shared broadly cultural and historical experiences
differing decisively from those of more distant peoples. Thus the diverse
peoples of India cherished in common the Sanskritic traditions, and those of
Europe the Greco-Latin. In the field of pre-Modern Afro-Eurasian history a
number of such broader cultural complexes have existed, not necessarily
covering without remainder the whole field of citied and lettered life, yet by
and large dominating the more local cultures.

There are many ways of grouping into 'civilizations' what is in fact an end-
less chain of interrelated local cultural life. We must know why we make the
selection we do. Often one may make alternative combinations according to
what questions one is concerned with. Thus the civilization that united the
lands from Nile to Oxus in the Islamic period could be regarded, for some pur-
poses, as no independent cultural body but simply the latest phase in a long-
term Irano-Semitic civilization continuous from the time of the ancient
Sumerians. The cultural traditions associated with Islam in India, then,
would be regarded as forming part of an equally continuous Indie civilization,
complicating it and relating it to the lands from Nile to Oxus, yet still re-
gionally bound. This makes obvious sense on the level of everyday life. Even
on the level of high culture, Islamic faith, for instance, is part of a longer tradi-
tion. In particular, the creative events at the founding of Islam were them-
selves part of an ongoing monotheistic tradition. They took place within the
setting formed by the dialogue that was working through the implications of
the ancient Hebrew prophetic discoveries; they formed a response to challenges
presented historically in that dialogue at the point when Muhammad was
drawn into it. In the Islamic dialogue, the same basic monotheistic commit-
ment persisted, though the particular allegiance was new. The like was true
of many other facets of high culture even though new languages were used.

If one is concerned primarily with socio-economic institutional evolution,
conscious or unconscious, especially with that of the more local social units,
such regional groupings may provide the most intelligible fields of study.
(Grouping by region, itself, would offer a number of alternatives: for instance,
instead of distinguishing between Europe on the one hand and the 'Middle
East' from Nile to Oxus on the other, one might, especially in some periods,
distinguish a Mediterranean region from a region centred round the Iranian
plateau.) But if one is concerned primarily with the more conscious commit-

0.2 GENERAL PROLOGUE

ments of human beings and with their public actions in the light of those
commitments, as historians traditionally have been, a grouping by traditions
of high culture will be more relevant than one by socio-economic regions.

When we speak of a great civilization, we mean above all a consciously cul-
tivated human heritage — and only secondarily a collection of folkways or of
sociological raw data. In the study of any culture, of course, intellectual,
economic, artistic, social, political aspects all have their role; ruling classes,
peasant villagers, city artisans, bands of vagrants, all must be taken into
account in interpreting it. But in studying a given civilization our first interest
is in those aspects of culture that have been most distinctive of it; that have
been most interesting and humanly significant in their variation within its
own sphere of space and time as well as in their diverging from other forms of
culture. During much of history, at least, this has meant the artistic, philo-
sophic, scientific life, the religious and political institutions, in general all the
more imaginative activities among the more cultivated of the population. It is
in terms of these aspects of culture that we do commonly distinguish the great
civilizations from one another. And in the long run, these aspects can be de-
cisive even for the common people who are scarcely aware of them.

This fact is especially relevant among the Muslims. Wherever it went,
Islam entered into the local cultural complexes carried by local ethnic groups,
as one tradition among the complex of interdependent traditions which go to
make up a local culture. But these local cultures might have nothing else in
common. It is not on this level that the wider culture {religious and non-
religious) associated specifically with Muslims was articulated. Such a wider
culture was carried primarily on the level of 'high culture', rather than folk
culture: that is, on the urban, literate level and more particularly on the level
of the cultivated circles who were direct or indirect beneficiaries of the land
revenues and who participated in the large-scale institutions that imposed a
social order wider than that of family or village unit. 1 It is this culture —
including, of course, its religious components — that we are to study in this
work. A civilization in this sense will normally be defined by a continuity of
lettered traditions: that is, of literature in the widest sense of the word, in-
cluding (for instance) religious or scientific literature. Hence our field, in
studying the civilization associated with the Muslims, will be delimited not by
geography — as, for instance, the culture of the 'Middle East' in one or another
acceptance of that term — but rather by the lines of development of the high
cultural traditions, wherever they lead us. 8

7 Edward Shils, in 'Charisma, Order, and Status', American Sociological Review, 30
(April 1965), 199-213, suggests how much the high culture (his 'central value system')
rests on the charisma attributed to it because it represents an inclusive social order felt
to be just.

8 What the civilization is that we associate with Muslims, and what its position is in
world history, have been discussed from many viewpoints; but rarely in an adequately
comprehensive world-historical perspective or with sufficiently flexible categories. Among
the best discussions have been three which form a sequence. Carl H. Becker, in Vom

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 93

Such a definition of our field can have subtle consequences. It makes for a
special way of viewing religion, for instance. In the historical developments
that maTk a civilization so defined, religion almost inevitably plays a key
role; but not necessarily so much because of the inertia of folk habits as be-
cause of its place in the consciences of a concerned minority. For the ideals of
a minority can be specially seminal on the level of high culture.

Any civilization, as a delimitable complex of cultural traditions, has been
constituted by standards of cultural valuation, basic expectations, and norms
of legitimation, embodied in its traditions. In the high culture, these are carried
partly in lettered traditions directly, and partly in other traditions, such as
social and artistic ones, associated with the lettered traditions. It is the more
far-reaching standards of legitimation that have served as the most depend-
ably persistent cultural traits, endowing a civilization with such cultural inte-
grality, such distinctive style, as it has had. What sets off most clearly one
civilization in our sense, from another, then, is not so much any general stock
of cultural ideas and practices — easily borrowed from one people to the next —
as these formative ideals. Such a cultural heritage has been carried not only by
all of the upper, educated classes, but even, to a lesser degree, by still wider
sections which have absorbed something of its outlook, down to the ordinary
peasants. But within this mass, a much smaller group has played a special
role: those who have taken the more articulate and far-reaching ideals of the
heritage as a personal responsibility, which they must themselves realize. This
concerned minority for whom cultural or spiritual ideals are a major driving
force are not usually the men of immediate power. But at every crossroads,
they are the men of cultural initiative — it is within the framework they have
clarified that new cultural choices must commonly be made.

At least in pre-Modern times, the most important focus of persistent cul-
tural ideals has often tended to be in religion. In religion, the impact of the
creative, revelatory events has tended to stand out most strongly from the
continuing dialogue in which their implications were being worked out; hence
religion could provide unusual continuity in the dialogue itself. Moreover, even
more than in the aesthetic or the political spheres, the circle of responsive con-

Werden and Wesen der islamischen Welt: I slamstudien , Vol. I (Leipzig, 1924), Part I,
'Zur Einleitung', brought out, among other things, that the pre-Modern Christian and
Muslim societies lived by largely common cultural resources. This point was developed in
a new direction by G E. von Grunebaum, in Medieval Islam (University of Chicago Press,
1946 [2nd ed. 1953]), who stressed the parallelism of their world views. Jorg Kraemer, in
Das Problem der islamischen Kulturgeschichte (Tubingen, 1959), has recently reviewed the
questions suggestively, attempting to balance the 'Hellenic' character of the culture with
other elements. Unfortunately he, like most scholars, still presupposes untenable notions
about a fictitious 'Orient', which lead much of his argument astray. He fails to see that
the data he cites to support the 'Oriental' character of Islamicate culture illustrate
instead the indivisibility of the Afro-Eurasian historical complex as a whole, which
included the Occident.

For further development of my own thoughts on defining civilizations, see the
section on historical method in the Introduction, especially the subsection on civiliza-
tions.

94 GENERAL PROLOGUE

frontation that spreads out from religion tends to be comprehensive of all life.
A new aesthetic impulse may affect science in some degree, or have economic
repercussions as fashions change. A new political impulse can reach further,
can carry in its train economic decisions, aesthetic ideals, or the very tone,
sometimes, of private life caught up in the pride of a new allegiance. A fresh
sense of impact from whatever it may be in the cosmic presence that is seen as
transcending the natural order may be felt strongly by only the concerned
minority. But if these take it seriously, it can touch every point in the natural
order of human affairs: it can reorient people's aesthetic sense, their political
norms, their whole moral life, and with these everything else that can be seen
to matter. A religious commitment, by its nature, tends to be more total than
any other. Perhaps especially among Muslims, religious vision has often proved
decisive at just the points that are historically most interesting. Moreover,
that vision proved sufficiently potent to ensure that Muslims formed a single
great civilization of their own.

Among Christian or Buddhist peoples, religion has indeed been very central
also. But it has informed the culture of Christian Occidentals and of Christian
Abyssinians, for instance, almost entirely in isolation from each other, so that
there is no single civilization associated with Christianity. Nor is there one
civilization associated with Buddhism. But — despite the vaster areas covered
— those who participated in the tradition of Islamic faith, so far as they
developed any culture of their own at all, never lost contact with each other:
their cultural dialogues were always intermeshed. The bonds of Islamic faith,
indeed especially the irrepressible transcendent ideals implied in the root
meaning of islam, with their insistent demand for a godly transformation of all
life, have been so telling in certain crucial aspects of the high culture of almost
all Muslim peoples that we find ourselves grouping these peoples together
across all their different regions, even apart from considering other facets of
high culture. Islam offered creative impulses that ramified widely throughout
the culture as a whole, even where it was least religious. It is largely around
the central Islamic tradition that the concerned and the creative built and
transmitted a common set of social and, above all, literary traditions; these
were carried in many languages but looked largely to the same great classics,
not only religious but secular, and especially to the norms which they express,
applicable to all aspects of life. Thus Islam helped to knit together peoples who
otherwise might have remained remote, or have drifted apart if they were
close to begin with. Through the greatest diversity of forms (as the chart
giving an overview of the history may suggest), these traditions (religious and
otherwise) have maintained a decisive continuity. Hence in studying these
peoples there is special urgency for studying as a body, and hence primarily
on the level of the high culture, the civilization given definition by the lettered
traditions in which Islam held a central place. 9

* The student will find that differences in scholars' notions of what it was they were
studying have left a profound impress upon the works which anyone must read if he is to

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 95

In studying the history of Muslims, obviously, we need distinct terms for
the religious tradition on the one hand and for the more inclusive civilization
on the other. Unfortunately, we have not had such terms in the past. The
terms 'Islam' and 'Islamic' have often been used in both senses. But these two
terms are clearly appropriate only to the realm of religion. If we speak in this
work of 'Islamic' art or literature, then, we will be referring to religious art or
literature within the traditions of Islamic faith, in the same sense as we refer
to 'Christian' art or literature. We will require a different term for the cultural
traditions of the civilization at large, when we are not restricting our reference
to religion. The various peoples among whom Islam has been predominant
and which have shared in the cultural traditions distinctively associated with
it may be called collectively Tslamdom', as forming a vast interrelated social
nexus. The distinctive civilization of Islamdom, then, may be called Tslami-
cate'.

The civilization could have been given some other name than one derived
from Islam; in fact it has, in some contexts, been referred to appropriately as
the 'Perso- Arabic' civilization, after the two chief languages in which it has
been carried. But because of the pre-eminent role played in it by Islam and by
Muslims, it has most commonly been called the 'Islamic' civilization. It will be
convenient to retain such a usage here, only adding the double ending (-icate)
to avoid an ambiguity that has proved all too common. In some cases, the
distinction is unimportant, and the choice between the terms 'Islamic' and
'Islamicate' may be a matter of emphasis. But on occasion it is essential to
point up the distinction between those traditions associated relatively closely
with the act of islam and its spiritual implications, and those traditions that
were associated with Islam more indirectly, through forming a part of the
overall civilization in which Muslims were leaders. The form 'Islamicate' has
the advantage of being almost self -defining: if it appears in a context where it
is contrasted to 'Islamic', it is clearly not just the same as 'Islamic' but does
relate somehow to what is Islamic. This is approximately the effect intended. 10

ISLAMICATE CIVILIZATION AS HUMAN HERITAGE

The Islamicate culture of the past has, of course, been very important in
influencing the present condition of mankind. It is naturally the major in-
fluence from the past among the widespread Muslim peoples of our time;
moreover, it has had notable effects, for good and for ill, in still wider areas,
such as much of India and of Europe, where Muslims once ruled. The civiliza-
tion remained actively creative within its own terms until the moment when a

pursue even casually any given line of interest in the Islamics field. In the paragraphs on
the history of Islamics studies in the Introduction is a brief sketch of the major orienta-
tions and biases which the reader should learn to be aware of and, if necessary, discount.
10 For an explanation of my choice of the terms 'Islamdom' and 'Islamicate', see the
section on usage in Islamics studies in the Introduction.

Late Sasa.nl and
Primitive Caliphal
periods, c. (485)-692

High Caliphal
Period, c. 692-945

Earlier Middle
Islamic Period,
c. 945-1258

Later Middle
Islamic Period,
c. 1258-1503

Period of Gunpowder

Empires,

c. 1503-1789

Modern Technical
Age, c. 1789-present

Overview of the History of Islamdom

The intrusion of Islam into Irano-Semitic society and
the genesis of a new social order. In Iran, in the Fertile
Crescent, in Arabia, the way was being prepared, as it
turned out, for the new order. First came the shaking
up of the old Sasani political order; but the central
event of the period was the advent of Muhammad
and his followers' rise to power from Nile to Oxus and
even beyond.

The first period of Islamicate civilization proper :
A classical civilization under the Marwdni and earlier
'Abbasi caliphates. Islamicate society formed a single
vast state, the caliphate, with an increasingly dominant
single language of science and culture, Arabic. The
Islamic religion was being given its classical formula-
tion; Muslims, Christians, Jews, and Mazdeans were
renovating and weaving together the lettered traditions
of several pre-Islamic backgrounds into a creative
multiple flowering.

Establishment of an international civilization spreading
beyond the Irano-Semitic areas. The great expansion of
Islamicate society was based on a decentralization of
power and culture, in many courts and in two major
languages, Persian and Arabic. Unity was maintained
through self-perpetuating social institutions which
outgrew the caliphate and encouraged high-cultural
sophistication and a synthesis of the lettered traditions
that had been developed in the High Caliphal Period.
The age of Mongol prestige : crisis and renewal in the
Islamicate institutions and heritage. Despite
devastation and conquest of the central Islamicate
lands by a vigorous pagan movement, the Islamic
norms reimposed themselves and hemisphere-
wide expansion continued. The Mongol challenge
launched a new political tradition and new horizons in
high culture in the central areas, forming a Persianate
culture from the Balkans to Bengal and influential
even more widely.

Flowering of Persianate culture under major regional
empires. The political and cultural impetus of the
Mongol age was developed in regional empires with
relatively regional cultures, especially in three : one
primarily European, one centered in the old Islamic
lands, one Indie. It was the height of Islamic material
world power. The aesthetic and intellectual creativity
and prosperity faded, however, before the new
Occident in the course of a basic transformation.
The Islamic heritage caught up in the Modern techni-
calistic world. Under the impact of a new world order
carried by the Modern West, the world-historical
conditions of the Islamicate civilization have
disappeared. Instead of a continuing comprehensive
society, we have a heritage which several peoples
share within a wider social order where Muslims form
a minority, a minority disadvantaged by just those
events which, creating the new order, brought
prosperity to the new West.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 97

transformed Modern Occident put all pre-Modern heritages in question. Until
that time the Muslims and their society played a pivotal role in world
history as a whole, both negatively and positively; hence in almost every part
of the Old World — and in the New World cultures derived therefrom — even
where Islam never prevailed, at least some elements of the local culture have
been traceable to Islamicate sources. Through its manifold influence on the
Medieval Occident, particularly in the realms of science and technical skills,
the older Islamicate culture had a significant share even in the far-reaching
cultural transformations which the Modern Occident has introduced to our
present world.

Important as it has been for its effects on the course of history, Islamicate
civilization may be still more important for us as illustrating the evolution of a
civilization as such. As we follow the traditions of that civilization in their
many forms and many spheres of activity, from a time before there could be
said to have been an Islamicate civilization at all, we are presented with an
instructive instance of a major civilization as an evolving, historical pheno-
menon. Every degree of integration and disintegration, of freshness, maturity,
decadence, and revival is illustrated in the most varied historical patterns.
Moreover, the roots of Islamicate civilization are largely the same as those of
Occidental civilization: the urban commercial tradition of the ancient Fertile
Crescent, the Hebrew religious challenge, the classical Greek philosophical and
scientific culture. Hence for Westerners (and for all who at least partly share
now in the Occidental heritage), the Islamicate forms a sister civiliza-
tion, like yet very different; acquaintance with it can throw a special light,
by way of comparisons and contrasts, on Occidental civilization in particular
as well as on the nature of civilizations as such.

Comparisons between the Islamicate culture and the Occidental are inevit-
able and very worthwhile. But one caution must be kept in mind. It is not
ordinarily legitimate to compare pre-Modern Islamicate institutions and cul-
tural patterns with those of the Modern West, and to treat that comparison as
if it were primarily one between different peoples. Such a comparison is more
likely to be one between ages. In recent centuries, enormous changes have
supervened in the Occident; changes which produced their counter-effects al-
most immediately, and are now having their analogues elsewhere, in the Muslim
and in other non- Western countries. Serious comparisons between the Muslim
peoples and the West should be made with this fact in mind. Pre-Modern
Islamicate ways can be compared with those of the pre-Modern Occident; and
those aspects of society in Islamdom in which the Modern changes have taken
positive effect can be compared, within measure, with the Modern West. But
it is illegitimate to regard as 'Western' — in contrast to 'Islamic' — such traits
as clock sense among workers, or democratic social expectations, or even
subtler characteristics often cited; for these traits were mostly as absent in
the pre-Modern Occident as in pre-Modern Islamdom, and may in the future
prove as congenial to Muslim peoples as they now seem to be to Western,

98 GENERAL PROLOGUE

But from a more deeply human point of view, perhaps neither the far-
reaching historical effects of Islamicate civilization, nor its value as illustrating
the nature of civilization in general, is so important as is the inherent marvel
of what it built, and even of what it dared try to build. Our fundamental pur-
pose must be an understanding of the human achievements of the civilization
in their own terms. It is with this intention that our field of study has been
delimited here, and the place of everything in it assessed. The place of the
civilization in the world-historical chain of events, and its usefulness as an
example of what culture can be, come out naturally and necessarily as we try
to understand what is special to the human endeavours that have been tied
together by the presence of the Islamic vision. In studying Islamicate civiliza-
tion, we will be concerned with society and culture as the context in which
concerned individuals have worked, especially Muslim individuals, and as the
handiwork which exhibits in varying ways the intended and unintended re-
sults of their work and of their vision. 11

Even in terms of evident relevance for current personal life among Modern
humans, the Islamicate heritage is rich. Its visual arts, for instance, include
surely the greatest ever known in which the element of sheer visual design
could be given priority over all other considerations. Its literatures, richly
unmatched in their most distinctive genres, are perhaps unparalleled in—
among other things — their mastery of the esoteric as a dimension of human
experience. Its philosophical and scientific and religious thought has not
merely made a lasting contribution to subsequent knowledge; much of it pre-
sents continuing points of enduring challenge. As we watch the unfolding of
the civilization as a whole we will gain, at the same time, essential background
for appreciating the monuments of Islamicate culture which can still enrich
our understanding and our life. For it is only in their total context, in the
setting of the developing cultural institutions which formed them, and of the
hopes and fears they embodied, that the monuments can come fully alive for
us: works of architecture and painting, literary masterpieces, philosophical
systems, expressions of religious insight, and, above all, living religious and
social institutions among millions of mankind. Perhaps the latter sort is the
most important of all the monuments of Islamicate civilization, if only because
the Islamicate society represents, in part, one of the most thoroughgoing at-
tempts in history to build a world-wide human community as if from scratch
on the basis of an explicitly worked out ideal.

But important as is the heritage of Islamdom to us as presenting resources
for our current cultural ventures, the great human venture which Islam has
been is even more important, for Modern mankind, as a venture: as it was in
itself then. If the Modern Technical Age is to remain human, it cannot over-
look the trust that our ancestors have left with us. Our past cannot be mere

1 ' For a fuller presentation of the historical viewpoint which I have used in this work,
see the section on historical method in the Introduction, especially the subsection on
historical humanism.

THE ISLAMIC VISION IN RELIGION AND CIVILIZATION 99

matter for a more or less curious utilitarianism, like iron deposits, say, on the
moon. Islamicate culture is supremely important because it represents the
highest creative aspirations and achievements of millions of people. Whoever
we are, the hopes, the triumphs, and the failures too of any human beings are
properly of concern to us; in the moral economy of mankind they are also our
own hopes and failures. In studying and sharing in them we know ourselves
better, understand better who we truly have been and are, we human beings.

BOOK ONE

The Islamic Infusion:

Genesis of
a New Social Order

Isaiah answered. I saw no God, nor heard any,
in a finite organical perception; but my senses
discover 'd the infinite in every thing, and as I
was then persuaded, &• remain confirm' d, that
the voice of honest indignation is the voice of
God, I cared not for consequences but wrote.

— William Blake

1 1 i

The World before Islam

The Islamicate was unique among the great civilizations of its time in failing
to maintain the earlier lettered traditions of its region. Elsewhere, the master-
pieces of the first millennium b C continued to form the starting point for intel-
lectual life. Right up to Modern times, the classical Greek and Latin (and even
ancient Hebrew) masters were read in Europe, their contemporary Sanskrit
and Prakrit masters in the Indie regions, the Chinese in the Far East. In
Islamdom, on the contrary, the Semitic and Iranian literatures of the pre-
ceding periods were gradually replaced by Arabic, and later Persian, during
the early centuries of Islam. Except in special little groups they died out,
relatively little surviving even in translation. Indirectly, elements of the old
lettered traditions persisted strongly in the new; but the great ancient works
were mostly unknown to Muslims in the original or in translation. Instead, the
Muslims developed their own classical models afresh. On the conscious literary
level where the consciences of cultivated persons are engaged, the coming of
Islam, then, marked a breach in cultural continuity unparalleled among the
great civilizations we have come to know; a breach which can help to produce
an impression of youthf ulness — or of immaturity — on observers more at home
in civilizations with a longer explicit heritage. 1 The breach with the older
regional heritage was later emphasized still more when the Islamicate civiliza-
tion, again uniquely in its time, became so widely dispersed over the hemi-
sphere that it ceased to be associated exclusively with a single region and
became dominant even in the heartlands of the older Greek and Sanskritic
traditions.

Yet the Islamicate society was not only the direct heir, but in significant
degree the positive continuator of the earlier societies in the lands from Nile
to Oxus. By geography and in point of human and material resources, it was
ultimately heir to the civilized traditions of the ancient Babylonians, Egyp-
tians, Hebrews, Persians, and their various neighbours; more particularly, it
was heir to the traditions expressed in the several Semitic and Iranian lan-
guages cultivated during the centuries immediately preceding Islam, tradi-
tions which in turn had built on the more ancient heritages. In their more
routine dimensions, life and thought did not greatly differ in the earlier Islamic

1 For a discussion of this and other circumstances — especially the accidents of
Western scholarship — which have helped produce in outsiders a sense of aridity in the
Islamicate civilization, see my 'Islam and Image' (a discussion of Muslim iconophobia) in
History of Religions, 3 (1964), 220-60.

103

104 THE ISLAMIC INFUSION

centuries from what they had been in the later pre-Islamic ones. Numerous
details in the culture of the Islamic period — art motifs, social customs, the
presence of minority religions such as the Christian — make sense only in terms
of the earlier cultures which first produced them. What is more, the Muslims
inherited also in large measure the problems, the opportunities, and the temp-
tations of their ancestors in the region. Even those aspects of the civilization
which were most strikingly new — for instance, the Muslim religion itself —
were formed in the context of the earlier Irano-Semitic traditions. The goals
to be set, the norms to be abided by, had been adumbrated long before.

Hence the achievements of the Muslims, in the unexpressed implications of
their writings or their art, in the deeper problems solved in their institutions,
often presuppose the continuing lifeways formed before Islam in the whole
region; to savour these achievements, even to assess their uniqueness, one
must recall the motive forces of that earlier life. What differed under Islam
was largely the relative weighting of different elements in the culture, the
balance among them. In working out that new balance, the impulses which
formed Islamicate culture proved to be exceptionally comprehensive and self-
sufficient.

Even these impulses go back into pre-Islamic times, however. With the proc-
lamation of Islam in Arabia, or at latest with the subsequent Muslim con-
quests, a new subculture, a new complex of cultural traditions, appeared with-
in the existent Irano-Semitic societies. But this was not yet in itself the
Islamicate civilization, though its heritage later formed the decisive element
in defining that civilization. As to the substance of social traditions — basic
expectations, knowledge, and even taste — others of the many heritages which
went into forming the civilization were commonly more decisive in it than the
heritage of the nascent Muslim community. The actual civilization, then, took
time to form. Even as to the Islamic contribution itself, it was only with time
that the developing Islamic traditions could penetrate into the various aspects
of the existing high cultural life of the times sufficiently for it to take on
identifiably Islamicate forms. But much that was to be associated explicitly
with the Islamicate civilization arose less from interaction with the Islamic
traditions themselves than from independent new developments within the
older traditions. Some of these were well launched long before Islam.

Accordingly, we must recognize an ill-defined period of gestation of the
Islamicate civilization, when its characteristic traditions were taking form
and being brought together. This period began long before (and ended a con-
siderable time after) the crucial event — the life of Muhammad — which marked
the beginning of the new subculture. Here we will try to trace, in this pre-
Islamic background, the developments that gradually sharpened within it in
the direction of Islamicate culture. We will begin by recalling certain long-
enduring overall social traits which have now very nearly vanished, but which
must be borne in mind at all times in appreciating any work of Islamicate
culture.

THE WORLD BEFORE ISLAM 105

The culture of agrarian-based society

We necessarily possess some image of society and culture in civilized lands be-
for the Modern Technical Age — an image usually influenced at least vaguely
by Karl Marx or Max Weber and their masters. In order to specify effectively
what was distinctive in the Islamicate development, I shall have to single out
points in pre-Modern social structure that were crucial to that development
as I see it in this work; and then to define a term that will (I hope) bring to
mind the particular complex of phenomena I find relevant — which does
not coincide exactly with those that Marx or Weber have brought into
view.

When men first built cities and extended an urban governmental authority
over the surrounding villages, they posed in a new form the dilemma between
social privilege and equal justice. In the valleys of the Euphrates, Tigris, and
Karun, and of the Nile, and in several neighbouring lands, the city dwellers,
especially the wealthier among them, enjoyed a substantial share of whatever
was produced in the countryside, beyond what the peasants themselves needed
so as to keep going. This was the 'revenue' of the lands. It was regarded as at
the disposal of whoever wielded power locally, and served to support those
who performed such essential functions as storing grain against the day of
bad harvests, maintaining internal order, defending the area against outside
predators, and, directly or indirectly, an increasing variety of other specialized
tasks ranging from the propitiation of natural forces to the importation of
exotic objects, such as metal, which all had need of.

Those who controlled the revenues patronized all that was refined in cul-
tural life above the level of village subsistence; and the quality of this culture
tended to depend directly on the material prosperity of its well-to-do patrons.
Thus it was the wealthy who patronized the fine crafts which, in the cities,
produced beautiful objects of leather and cloth and wood, as well as of bronze
and silver and gold, for ornamentation and for all everyday purposes; what-
ever came under the hand or the eye of the wealthy was a specially designed
work of art. The craftsmen passed on their methods from father to son; the
methods included as much the aesthetic standards, and even the particular
aesthetic forms, as they did the physical technique, from which indeed the
forms were inseparable. The excellence of the result depended largely on the
quality of materials used, the amount of time the craftsman could spend on a
piece, and the degree of discipline of the craft tradition; all these depended in
turn on the patrons. When the revenues from the land were great, when the
wealthy who received them could use them in security to gratify their tastes,
and when those tastes in turn had been cultivated through lifetimes of high
standards, the craftsmen were able to develop their skills and put forth their
best efforts. If, on the other hand, times were troubled, and for some reason
the wealthy could not collect much revenue from the peasants, or could not

106 THE ISLAMIC INFUSION

use it freely for their private tastes, the quality of craft work was likely to
decline. 2

It was in the cities likewise that more monumental art was produced — fine
massive buildings, and the statuary and other carving and painting that went
with them. Enormous effort was commonly put into temples which, as expres-
sions of the honour paid by the community to the gods, were to the interest of
all, or at least all in the cities; commonly the temples represented the best that
men's resources could achieve. At the same time, the homes and courts of
rulers and of well-to-do individuals were likewise built as sumptuously as such
individuals could command. Finally, it was in the cities, and among those re-
leased by wealth or office from the everyday labour of the peasant, that people
produced a refined literature of ritual and of myth and legend; a literature
which finally came to enshrine a sense of personal conscience in the face of the
cosmos. Such monumental literature, like the monumental architecture, was
often devoted to magnifying rulers also, whom indeed it was sometimes diffi-
cult to distinguish from the more natural Powers.

All these arts of civilization, then, were dependent on the patronage and
appreciation of a limited number of privileged persons in the cities. As in the
case of handicrafts, when wealth failed to concentrate peacefully in their
hands, standards of excellence declined. Their wealth, in turn, depended on
the subjection of the bulk of the population, especially the peasants.

Most persons who troubled to compare the state of the great and the lot of
the peasants, or other lesser beings, were content to observe that a mere peas-
ant, rude and uncultivated from his childhood, had all that was due him if
he had just enough to live on. Yet, very early, voices were raised in doubt.
One of the arts of civilization was the art of legally enshrined justice; a ruler
might pride himself not only on a magnificent palace, or on a magnificently
composed record of his awesome exploits, but also on a reputation as a giver
of just laws. As the cultivation of a personal conscience came to the fore
among the civilized arts, the pride of justice might be expected to loom ever
larger.

At first it was the temple that was the focus of whatever high culture there

1 It has seemed necessary to try to delineate, here and in later chapters, a number of
traits general to pre-Modern life. Increasingly, attitudes and presumptions of an earlier
age become strange to us; even circumstances that seemed self-evident as late as fifty
years ago have become unintelligible to younger readers unless they are specially re-
minded of their occasions. Yet all historical judgment is made through comparison, expli-
cit or implicit. If non-comparable items are used as the basis for such comparison, the
results are false. In Islamics studies a range of false assumptions has arisen about
politics, religion, literature, progress and decadence, and so on, derived from various
phases of Western experience on the basis of just such false comparisons. Questions have
been put, on the basis of supposed comparability to Modern Western situations, which
are irrelevant in the pre-Modern Islamicate context; conversely, truly pertinent questions
have too often been overlooked. To know what must be explained as special to Islamicate
society, we must recognize what was usual to pre-Modern society generally. Even scholars
have too frequently been misled on such points; other readers will have yet greater need
of being reminded of the broader context.

THE WORLD BEFORE ISLAM IO7

was. At the temples in ancient Sumeria, where urban life began in the fourth
millennium bc, the work of controlling the local flooding and providing for
the drought of the Mesopotamian alluvial plain was carried on under the
learned priests, who in turn disposed of the surplus. It was they who sent out
traders to bring in exotic goods necessary to the developing exploitation of the
plain, fertile but lacking in minerals and even stone. When disputes arose with
rival towns, perhaps over control of the trade, they organized the fighting
men. But then as warfare became more elaborate — each town trying to outdo
the others — military affairs and the general control of the town fell into the
hands of non-priestly specialists: kings and their dependents. The royal court
became a second focus of high culture alongside the temple, and was based
like it upon agricultural production. Its revenue, in whatever form it took it,
may be called taxes, which came chiefly from the land. Much more gradually,
at last, the traders too became independent merchants, doing business on
their own account and gaining enough profit to share, if more modestly and
indirectly than temple or court, in the revenue of the land. When this hap-
pened, rich merchants too became patrons of the arts and the market became
a third focus of high culture.

All three foci of high culture depended on the condition of agriculture. The
basis of temple and court was agrarian in that their wealth and power pre-
supposed chiefly arrangements concerning agricultural production. The mar-
ket depended on agriculture less directly than did temple or court, for the
traders brought goods from afar subject to other hazards than that of the
local weather, and (provided there were sufficient stored savings) sold their
goods in lean years as in fat. Yet, in the long run, the merchants too depended
on the state of agriculture and their profits presupposed the peasants' surplus.
Even when, as in Syria, mercantile city-states arose which depended pri-
marily on distant trading by sea and land, their trade depended so intimately
on the agrarian societies about them that both morally and materially they
too lived ultimately from the peasants. Even the pastoralists, including the
desert nomads, who depended on the agriculturists for much of their food and
goods, were part of the same social complex. Accordingly, the type of social
order which was introduced into the agricultural regions (and the areas depen-
dent on them) with the rise of cities may be called agrarian-based or (to be
more comprehensive) agrarianate citied society. (I say 'citied', not 'urban', be-
cause the society included the peasants, who were not urban though their life
reflected the presence of cities.)

We shall use the phrases 'agrarianate' society or culture to refer not just to
the agrarian sector and the agrarian institutions immediately based on it, but
to the whole level of cultural complexity in which agrarian relations were
characteristically crucial, which prevailed in citied societies between the first
advent of citied life and the technicalizing transformations of the seventeenth
and eighteenth centuries. The term 'agrarianate', in contrast to 'agrarian',
then, will refer not only to the agrarian society itself but to all the forms of

108 THE ISLAMIC INFUSION

society even indirectly dependent on it — including that of mercantile cities
and of pastoral tribesmen. The crucial point was that the society had reached
a level of complexity associated with urban dominance — in this sense, it was
'urbanized' — but the urban dominance was itself based, directly or indirectly,
primarily on agrarian resources which were developed on the level of manual
power: based on them not in the sense that all must eat but that (since most
production was agricultural) the income of crucial classes was derived from
their relation to the land.

The culture of agrarianate citied society can be characterized as a distinct
type in contrast both to the pre-literate types of culture that preceded it and
to the Modern technicalistic culture that has followed. In contrast to pre-
citied society — even to agricultural society before the rise of cities — it knew a
high degree of social and cultural complexity: a complexity represented not
only by the presence of cities (or, occasionally, some organizational equivalent
to them), but by writing (or its equivalent for recording), and by all that these
imply of possibilities for specialization and large-scale intermingling of differ-
ing groups, and for the lively multiplication and development of cumulative
cultural traditions. Yet the pace of the seasons set by natural conditions
imposed limits on the resources available for cultural elaboration; moreover,
any economic or cultural development that did occur, above the level implied
in the essentials of the symbiosis of town and land, remained precarious and
subject to reversal — in contrast to the conditions of Modern times, of our
Technical Age, when agriculture tends to become one 'industry' among others,
rather than the primary source of wealth (at least on the level of the world
economy as a whole).

We must recognize the great diversity within what we call 'agrarianate'
society, both as to the level of complexity it reached and as to the forms of
elaboration to be found in different areas. Fundamental changes took place
everywhere, especially during what we call the Axial Age (800-200 bc). At
that time, letters ceased being the monopoly of a priestly scribal class and
became widespread among a section of the bourgeoisie, and correspondingly
the character and pace of development of the lettered traditions changed; and
at the same time, the overall geographical setting of historical action was
transformed, being articulated into vast cultural regions spanning the hemi-
sphere among them.

Yet because of their common agrarian basis, pre-Axial Age and post-Axial
Age society everywhere shared certain basic ranges of opportunity for his-
torical action, and corresponding limitations on it. Productivity could be
multiplied through the proliferation of specializations in crafts and crops, made
possible by the centring of surplus resources in cities; but not beyond the
level at which the power of animals and of the natural elements could work.
Social organization could allow for a diversity of personal roles well beyond
that possible in pre-literate societies; but not such as to escape a severe social
stratification, in which the great majority remained excluded from regular

THE WORLD BEFORE ISLAM log

participation in political and high-cultural life. In particular, citied life implied
an accelerated pace of historical change, of those actions and events that
change the presuppositions of everyday life, to the point where the individual
could become conscious of such change and of the possibilities of his own
actions changing the life conditions of future generations. Among other things,
therefore, a social conscience became a more likely possibility. Yet in society
on the agrarianate level, basic change remained the exception and innovation,
in principle, an occasional matter, in contrast to the Modern Age, when in-
novation has become institutionalized. Historical and moral consciousness,
that is, escaped the anonymity of the locally immemorial and could even
envisage a vast geographical scene of action; but it could not escape the sense
that the past was per se authoritative, nor achieve a trans-cultural world-
wide perspective. 3

Cosmopolitan and mercantile tendencies in early agrarianate society

Within the limitations of cultural resources on the agrarianate level, change
was steady and far-reaching; and from very early such changes were inter-
dependent across a large part of the Afro-Eurasian landmass. Local societies
were less and less independent in their cultural development.

Agrarian-based citied society determined the main features of historical
development throughout an increasingly major portion of the Eastern Hemi-
sphere: that is, in those regions which together went to make up what the
Greeks called the 'Oikoumene', the 'inhabited quarter' of the world. The Greeks
conceived the Oikoumene as a geographically fixed area between Atlantic
and Pacific and between the equator and the uninhabitably cold north; but if
we define it in terms of the peoples the Greeks would have included explicitly
or implicitly, we must regard it as a historically developing complex. As
agriculture and the associated domestication of animals spread through the

- 1 The term 'agrarian' can properly refer to an agricultural order in which property
relations are disposed with reference to the sort of stratification and organization most
commonly associated with the presence of cities as key political and economic centres
and foci of historical initiative. So soon as cities developed, the agriculture — and also the
primitive commerce and industry — in their vicinity were thus subjected to urban in-
fluence; but always on an agrarian basis. The term 'agrarianate' seems comprehensive
enough, so understood, to include within itself both urban life of this sort, which pre-
supposed the economic resources concentrated by agrarian tenures as its mainstay {at
least if one sees any given urban life in its total economic setting) ; as well as such peri-
pheral economic forms as independent pastoralism, which also presupposed at least
agricultural society, and generally, in practice, the citied agrarian-based form of it.
Alternative terms seem all unsatisfactory for our purposes. 'Pre-Modern civilized' or
'pre-Modern citied' life fails to bring out the positive urban-agrarian character of the
social order itself.

The tendency in modern area-studies to lump all pre-Modern society as 'traditional' is
subject to many serious objections, which will appear abundantly in the course of this
work; not the least of them is that it fails to bring out the startling historical contrast
between conditions before and after the development of citied and lettered life. It also
presupposes a definition of 'tradition' that reduces it to immemorial prescriptive custom,
and thus drastically misrepresents the nature of culture on the agrarianate level.

110 THE ISLAMIC INFUSION

Afro-Eurasian landmass, all the peoples involved came to be historically inter-
related to some degree, tied into the trade network and subject to at least the
indirect impact of the historical developments that arose in the citied regions
in the older agricultural areas. All these peoples, even when south of the
equator, may be included as making up the historical Afro-Eurasian Oikou-
mene. This Oikoumene remained the setting of most historical life in the hemi-
sphere down to Modern times, when agrarianate society ceased to be the
determining form of society in the world at large, being superseded by Modern
technicalized society by the end of the eighteenth century. We may call the
period when agrarianate society was historically dominant, within the range
of the Oikoumene, the 'Agrarian Age' (lasting from the time of Sumer down to
the seventeenth and eighteenth centuries), in contrast to the Modern 'Tech-
nical Age' since the eighteenth century.

The history of all the peoples of this vast area was affected more and more
by the interrelations in the Oikoumene. Over the centuries, the areas within
the Oikoumene that were organized under urban rule expanded. New luxuries
were discovered and coveted, traders sought out more distant sources of
supply, and armies followed them to impose a more secure authority. New
techniques were developed in production — and in organization. Nowhere was
all this felt more than among the many peoples of the Fertile Crescent and
the neighbouring lands, who lived in the very midst of the Oikoumene.

These peoples, among whom Islam was to develop, were increasingly linked
together, even apart from wider contacts across the Oikoumene. At first, they
were the foremost example of the development of a cosmopolitan regional
high culture. Several peoples had come to use the cuneiform script of the
Sumerians, and among them all, the Sumerian lettered tradition (cultivated
especially by the priests) tended to be influential: to this extent, they formed
a single multinational civilization which can most conveniently be called the
'Cuneiform'. Meanwhile, political units became larger. Among Sumerians and
Akkadians, Hurrians, Hittites, Urartians, and many others, territorial king-
doms arose which learned effectively to control from a central capital the
politics of many cities; and then great empires arose, in which even such
larger nationalities were mingled and sometimes even dispersed. Under the
relatively restricted sway of the Hittites, then under the widespread empire of
the Assyrians, the populations of whole cities were transported from one land
to another for reasons of state. People came to have a wider perspective on
matters taken for granted in local societies. They could see more readily how
limited were local assumptions in craftways and in political patterns and even
in points of conscience. The cosmopolitanism of the future Irano-Semitic cul-
tural tradition was being launched.

The consequences of such a development were expressed in a shift in social
alignments. In the eighth and seventh centuries bc, the Assyrians were per-
fecting their imperial power by a combination of horsemanship, of cruel ter-
ror, of moral earnestness, and of administrative efficiency. It was the admin-

THE WORLD BEFORE ISLAM III

istrative efficiency that was probably most decisive, for it spread a single set of
high-cultural standards everywhere. Local traditions were perpetuated — ■
most notably in the worship of the gods; but increasingly they receded before
common traditions expressed in a new common administrative language,
Aramaic. Unlike earlier languages, Aramaic did not share the priestly Cunei-
form tradition, which from this time gradually lost its power; it was, however,
Semitic, like the most important of the Cuneiform languages. It became the
language (especially in the Fertile Crescent) first of the merchants and of city
life generally; from there it came to be used by the clerks at the courts (if not
in the temples) — and at last even among the peasantry. Other languages con-
tinued for some centuries as local rural dialects (and as the medium of religion
and science among the learned); but the more dominant elements of the
Fertile Crescent and the nearby highlands were becoming, with one regional
layman's language, substantially one diversified people, and shared many
common expectations.

It was not only in language that the growing power of the merchants and
clerks appeared. Even the Cuneiform literatures of the time reflect a growing
sense of personal individuality which most probably catered to the tastes of
the market more than to either temple or court. Perhaps a symptom of the
mood of many people of the time can be seen in the striking development of
astrology in this period. The movements of the stars had long been studied in
Babylonia for their bearing, among other things, on the fate of kings. There
developed, beginning about this time, in the Cuneiform Babylonian language
of the priests (who still monopolized such matters), a more systematic, 'scien-
tific' cultivation of mathematical astronomy. But by the end of the Axial Age
it was directed no longer merely toward understanding the fate of kings, and
so of their kingdoms at large, but also toward understanding the fate of ordin-
ary individuals (presumably, wealthy ones), as individuals and apart from the
common destiny of peoples.

The florescences of the Axial Age and the articulation of cultural regions

The old Cuneiform lands with their new practical Aramaic were not alone in
these tendencies. As, with trade, the citied regions in the Oikoumene had con-
stantly spread, the market had become a major focus of high culture also
elsewhere. In other areas too, scholars became concerned with individual
destiny. For partly unknown reasons, the different citied regions of the Oikou-
mene, even when there was no direct commercial contact (as between China
and the Indo-Mediterranean regions) developed along parallel lines. At
any rate, as is well known, toward the middle of the first millennium bc
everywhere men grew prosperous and new social and economic ways pro-
liferated, especially in matters mercantile; for instance, coinage, as a means of
ready financial exchange under complex conditions, seems to have been
developed roughly simultaneously, toward the seventh century, in Anatolia,

112 THE ISLAMIC INFUSION

in northern India, and in China. In the same widely distant parts of the
Oikoumene appeared unexampled works of intellectual creativity which
proved decisive for all subsequent high-cultural life. Cumulatively, all this
amounted to great bursts of creative and many-sided cultural innovation,
launching many new traditions of high culture: that is, to unparalleled cultural
florescences. We may adopt for this period, roughly 800-200 bc, the term
used by the philosopher Karl Jaspers, 'Axial Age', on account of its subse-
quent historical importance. 4 The age can be called 'Axial' not merely be-
cause — as the chart of the place of Islam in world chronology shows — it
comes at the middle of the citied Agrarian Age, but because it resulted in an
enduring geographical and cultural articulation of the citied zone of the
Oikoumene into regions.

One of the most significant outcomes of these florescences was, in fact, the
establishment of new and comprehensive complexes of high-cultural tradi-
tion, such as the Cuneiform had once formed. These were carried now in lay-
men's languages: the Aramaic was perhaps the first but also the least full}'
developed of these. The areas where the new lettered traditions developed
came to form core areas: areas within which the greater part of all new high-
cultural development was to be found thenceforth, or at least that which was
at all widely propagated. Among them, these core areas tended to give a lead
to the high-cultural life of the whole Oikoumene during the rest of the Agrarian
Age.

At the expense of a certain schematizing, Afro-Eurasian citied history
thenceforth can be presented as the interdependent and more or less parallel
development of four major complexes of civilized traditions: the European
complex, with its core area from Anatolia to Italy along the north of the
Mediterranean Sea, and with Greek (and Latin) as classical languages; the
complex in the region from Nile to Oxus, centred in the Fertile Crescent and
the Iranian highlands, and carried chiefly in a series of Semitic and Iranian
languages; the Indie, in the Indian area and the lands to the southeast of it,
with Sanskrit (and Pali); and the Far Eastern, in China and its neighbouring
lands. 5 As these regions were all in contact, there was much mutual influence
and even sharing of common heritages, for instance in commerce and art, in
religion and in science.

Still more important, perhaps, than simple borrowing among the regions
was their common historical context. The culture of each of the regions was
constantly expanding its sway beyond its original centres; thus jointly they
extended in all directions, millennium by millennium, the overall field of

1 I have found it necessary to use a number of new terms in this work; and other terms
I have denned in a way alien to the daily newspaper (e.g., such a geographical term as
'Syria', which refers to the whole area between the Sinai peninsula and the turn of the
Anatolianpeninsula) . For a defence of my policy, see the section on revisionism in scholarly
terminology in the Introduction.

5 For an explanation of my usage of such general terms as 'European' and 'Indie', see
the section on usage in world historical studies in the Introduction.

THE WORLD BEFORE ISLAM

113

The Place of Islam in the Chronology of World History

f IO,<

7000 isc(?) Early village farming communities

Agrarian Age

(within the Afro- ^ 3000 bc(?)

European Oikumene)

Technical Age

Cities

► Pre-Axial Age

800 bc ■<

► Axial Age

200 bc <

600 CE

1800 CE
Present

Islam

> Post-Axial Age

114 THE ISLAMIC INFUSION

mercantile endeavour and of civilized history. All shared in a cumulatively
increasing geographical network of commercial and cultural interchange and
in a cumulative multiplication of techniques of all kinds. The common his-
torical context was highlighted from time to time, moreover, by a sequence of
important events impinging upon most or all regions. The development of
Muslim power was to be one such event. In sum, the whole Afro-Eurasian
Oikoumene was the stage on which was played all civilized history, including
that of Islamicate civilization, and this stage was set largely by the contrasts
and interrelations among the great regional cultural complexes. And no region
was more exposed to the consequences of all these interrelations than the cen-
tral region of Cuneiform heritage.

These regional cultures were mostly set apart, in the Axial Age, through the
rise and common regional cultivation of one major lettered tradition in each.
It is the literary tradition associated with Confucius, Lao-tze, and their suc-
cessors that most specifically defines the Chinese- Japanese Far East (of
Eurasia), as such. From Thales in Anatolia, Pythagoras in Italy, Socrates,
Plato, and their associates, stems the classical heritage of the European re-
gion, eastern and western. From an early age, peoples of the hinterlands there,
Lydians or Etruscans, were drawn into the orbit of the traditions carried by the
main commercial language of the coasts, Greek. In the Indie region, the age of
the Upanishads, of Buddha, and of Mahavira had a somewhat similar de-
cisiveness.

The Cuneiform cultural region, set in the centre of Oikoumenic cross-
currents, was exceptional. Endowed with the oldest high-cultural traditions,
and earliest to develop new ways, it did not, at that time, go on fully to
develop its own identity in the new manner. Rather, it was subject to in-
fluences from all sides. Within the region, as if awed by the aura of years, the
newer Aramaic continued to share the honours of a lettered tradition with the
already entrenched Cuneiform languages right through the Axial Age. Nor did
Aramaic develop a high tradition within its own sphere based on major
Aramaic masterpieces. The region where Islamicate culture was to be formed
can almost be defined negatively: as that residual group of lands in which the
Greek and the Sanskrit traditions did not have their roots and from which the
European and Indie regions were eventually set off. For it is the latter two
regional cultures that were at first most strongly marked and integrated. * In
this sense, our region, in the Axial Age, consisted of those lands between the
Mediterranean and the Hindu- Kush in which Greek and Sanskrit had at best
only local or transient growths. (The map of the pre-Islamic world from India
through the Mediterranean shows its central position.) Yet here, too, a core
area and a core complex of traditions can be defined positively, as well, on the

6 Karl Jaspers, from whom I take the term 'Axial Age', almost failed to see the inde-
pendence of the region from Nile to Oxus at this period; and indeed it was always more
closely linked to the European region than to the Indie. See his The Origin and Goal of
History (Yale University Press, 1968).

THE WORLD BEFORE ISLAM 115

basis of regional creativity which took place in the Axial Age; though this was
not necessarily in Aramaic.

As in the other areas, it was concern with the private individual as personal,
as independent in some degree from the group of which he formed a part, that
increasingly exercised the great prophets who arose in the Axial Age, notably
Zarathushtra (Zoroaster) in Iran and the Biblical prophets among the
Hebrews. The prophets spoke to human beings in the name of a supreme and
unique God, not reducible within any image, visible or mental, but expressing
a moral dimension in the cosmos; they demanded unconditional allegiance
from each person to this transcendent vision. Zarathushtra and his successors
preached the duty of each individual personally to take part in a cosmic strug-
gle between good and evil, justice and injustice, light and dark; a struggle in
which finally light and truth must be victorious. On the personal level, the
individual's duty was to be expressed in purity of life; on the social level, in
maintaining a just balance among agrarian social classes. The Hebrew pro-
phets, in Palestine and later in Babylonia, called men and women to the love
of a Creator-god elevated above any nation, who would exact unusually
severe standards precisely of those he most favoured, but promised them in the
end compassion and fulfillment, when they should be prepared to worship
him in full moral purity. These prophets founded strong literary traditions in
old Iranian 7 and, above all, in Hebrew (and thence later in Aramaic), expres-
sing and developing their visions. Though, at the time, these traditions were
politically submerged and none found general intellectual acceptance, they
gradually became the most sacred heritages — developed in greatly diverse
ways — of an increasingly large proportion of the people of the Fertile Crescent
and the Iranian highlands; ultimately they quite replaced the Cuneiform as
well as any local heritages.

Accordingly, at least indirectly, the Axial Age launched the distinctive tradi-
tions also of the fourth region, what is commonly called the 'Middle East'.
This region may be defined for our purposes, roughly, as the lands from the
lower Nile valley to the Oxus basin, inclusive. 8 The core area of the region was
limited to the Fertile Crescent and the Iranian highlands, where Cuneiform
languages and Aramaic were used: from the Axial Age on through Islamic
times, it was books written in this limited area that were read beyond it, or

7 We have no dependable evidence of written texts in the Zoroastrian tradition before
Arsacid (Parthian) times; the transmission of the key compositions may have been oral,
though writing was known in eastern Iran in the Achaemenid period and the tradition
seems to have grown (and retained distinctions of language) rather as it might have if
written. It is not clear what was the role of a priest like Saena (who is mentioned pro-
minently as an upholder of truth) , either as a transmitter or as a relatively independent
presenter of moral challenge; but one must doubt whether Zarathushtra was the only
significant prophetic figure.

8 I prefer not to use the term 'Middle East* for this region, as it is misleading in more
than one way, and I refer instead to the 'lands from Nile to Oxus' or even to the 'Nile-to-
Oxus region'. I have explained my usage in the section on usage in Islamics studies in
the Introduction.

THE WORLD BEFORE ISLAM 117

institutions introduced there that were imitated elsewhere, and only rarely
the reverse. Here the literary traditions founded by the Iranian and Hebrew
prophets provided the most creative imaginative stimulus, giving rise subse-
quently to related traditions such as those of the Christians and the Mani-
cheans. Already within the Axial Age there was enough influence of one tradi-
tion or another to suggest a common name for the group. The languages of
these traditions were largely Semitic and Iranian, and we may call the re-
gional culture so represented, somewhat loosely, the 'Irano-Semitic'. 9

Personal conscience and the new society

In all four regions, the innovating thinkers shared certain orientations, per-
haps as a result of similar experiences during the mercantile expansiveness of
their time and because of the intermingling of peoples then (at least from the
Mediterranean to India). Very often, in any case, the cultural setting in which
they spoke was that of the market; it was rarely that of the temple. They all
spoke to the aroused individual conscience and founded their appeals on the
basis of individual, rational responsibility. Beyond that, however, the forms
in which they posed the problems, and naturally the answers they came to,
though there was much overlapping, differed radically in their most character-
istic presuppositions. The Indie thinkers concerned themselves above all with
exploring the resources of the self, developing subtle and comprehensive ways
of understanding and mastering its unconscious recesses; they sometimes
sought to transcend injustice {as in the Gita) by making irrelevant the terms
in which the question was posed. The Hellenic thinkers explored, above all,
external nature; even after Socrates' and Plato's emphasis on the moral
nature of the individual, the starting point of a philosophic synthesis normally
remained a system of nature as a whole; justice (as in the Republic) must be
founded in cosmic harmony, in which different sorts of individuals will play
different parts. The Irano-Semitic prophets analyzed neither the inner self nor
the outer world. If they analyzed anything, it was history itself; the Hebrew
writers, especially, developed a majestic awareness of moral progression in

' I prefer the term 'Irano-Semitic' for this cultural complex to 'Middle Eastern' (if
only because the latter phrase can be far too inclusive for this purpose); and to the
ubiquitous 'Oriental', which even in the most circumscribed German usage includes too
much. I see, indeed, no special need for a common term for all cultural traditions from
Nile to Oxus except insofar as the phrase 'Irano-Semitic' is in fact applicable; that is,
until the date when other traditions lost their autonomy and the whole regional culture
became, in fact, Irano-Semitic predominantly.

But I do not understand 'Irano-Semitic' in a genetic sense. I am not concerned here to
sort out — if that be possible — the remoter Iranian, Semitic, and Hellenic origins of
various cultural elements. Here I am concerned with how they developed as an inter-
related complex of cumulative traditions. For this purpose, the Semitic and Iranian
peoples formed a relative cultural unity over against the Greek and Latin-speaking peo-
ples. The relative amount of Hellenic elements in Gnosticism, for instance, interests me
here less than its development — in whatever language — within the overall context of
Irano-Semitic traditions.

Il8 THE ISLAMIC INFUSION

history (not, of course, of 'progress' as such). The prophets summoned the per-
sonal conscience to confront a cosmic moral order, which expressed itself in the
contingencies of social history; the struggle for justice was a matter of his-
torical action. For their more consistent followers, interpersonal, egalitarian
justice became the supreme value in civilized life, even, if need be, at the ex-
pense of the arts and accomplishments of luxury.

The developments of the Axial Age bore fruit in a new type of political
structure — and again earliest and least completely between Nile and Oxus.
The Assyrian military empire had collapsed, overwhelmed by the hatred of
many peoples for its cruelties. But after an interval, in the sixth century bc
even more of the lands from Nile to Oxus were absorbed into the Achaemenid
Persian empire, which united in peaceful interchange the many peoples from
the Aegean to the Indus. Under it the Aramaic merchants received full mer-
cantile freedom, and Aramaic became the chief administrative tongue even
throughout Iran. The Persians were careful to make no such violent breach
with old local traditions as had the Assyrians; perhaps they no longer needed
to, for the peoples had by now learned something of how to live in an inter-
national society. They preserved traditionally developed institutions every-
where so far as was still possible, without introducing important new social
principles; but they proclaimed with new insistence the supreme duty of
maintaining truth and justice, a duty imposed upon those privileged to rule.
Such an ideal was proclaimed even in their monumental art which, in contrast
to the Assyrian portrayal of generals destroying their enemies, commonly
presented the king in his court as lord of peoples. Many among these peoples,
in fact, like the Jews who were benevolently restored to their homeland,
learned to revere the Great King as guarantor of peace and prosperity against
interference by any lesser powers throughout the lands. 10

In each of the great regions there was a tendency, as from Nile to Oxus, for
political power to be built on an increasingly wide territorial basis, with the
consent and support of the cities of diverse peoples, who found a common
interest in the order such a power provided as their interrelationships became
increasingly complex. Empires were built, typically over all the core area, at
least, of each of the four regions, carrying not only military control but also a
degree of social and cultural unity. Such empires were made possible by the
considerable economic and social development that had taken place in each
region during the Axial Age. They tended to found themselves, at best, on
some elements of the best philosophic thinking of their respective regions. The

18 For an overview of ancient cultural life between Nile and Oxus, two volumes are
especially convenient: Henri Frankfort, The Birth of Civilization in the Near East
(London, 1968) and Henri and H. A. Frankfort, John A. Wilson, and Thorkild Jacobsen,
The Intellectual Adventure of Ancient Man (University of Chicago Press, 1946 — also in
paperback as Before Philosophy, Pelican Books). Gordon Childe, What Happened in
History (Pelican Books, 1942), is rather out of date but still a suggestive review of a
longer period. The master summary of present knowledge is now volume one of the
Unesco-sponsored History of Mankind (New York, 1963).

THE WORLD BEFORE ISLAM Iig

Maurya empire in northern India made use of Buddhist thought; the Ch'in-
Han empire of China was founded by anti-traditionalist theorists, and then
depended increasingly on Confucian standards; the Roman empire of the
Mediterranean built its law largely on Stoic assumptions about mankind, and
its social order generally on the ideals of Hellenic city life developed in the
time of the classical thought.

The regional tradition developing more weakly from Nile to Oxus suffered a
peculiar complication upon the fall of the Achaemenid Persian empire. Its
conqueror, Alexander, represented not Aramaic nor even prophetic but Hel-
lenic culture; that is, the civilization centred in the northern coastlands of
the Mediterranean. Hellenic merchants and mercenaries had played an in-
creasing, though subordinate, role already within the Achaemenid empire.
But for several centuries after Alexander, Greek culture and its carriers held a
more or less dominant position in much of the region. Throughout the central
area of the Irano-Semitic traditions, in the Fertile Crescent and the Iranian
highlands (as well as in Egypt), Greek cities, with Hellenic traditions and
expectations, flourished side by side with cities whose traditions continued
Aramaic and even Cuneiform patterns of life. In many areas the Hellenic
element was the more powerful.

This Hellenic strength came less from military force than from cultural
attraction. The Seleucid dynasty, which ruled after Alexander's death, was
explicitly Hellenic and depended upon cities formed in the first instance of
Greek colonists, scattered everywhere between Nile and Oxus. But the Par-
thian Arsacid dynasty, which soon succeeded it in the greater part of its
territories, was Iranian by tongue; yet it too was avowedly (at least at first) a
protector of Hellenic culture. Hellenic culture, minimally diluted with local
colour, enjoyed a near-monopoly of respect in the cosmopolitan big cities; the
old Cuneiform (and Hieroglyphic) traditions gradually died out, and not
Aramaic but Greek took their place as vehicle of the more highly cultivated
literature. The four or five centuries after Alexander, as far east as the Oxus
as well as in the Mediterranean basin, are reasonably called the 'Hellenistic'
age.

The eventual results of this superposition of Hellenic culture upon the Irano-
Semitic and Egyptian are still in debate. One thing is certain. The whole tra-
dition of natural science, centring on mathematics and astronomy, which had
been founded largely on ancient Babylonian data and which even in Seleucid
times was being developed simultaneously in Babylonian Cuneiform and
in Greek, gradually came to be indissolubly associated with the Hellenistic as-
pect of culture in the region; the fortune of natural science henceforth de-
pended upon that of the Hellenic elements there. This situation was perhaps
decisive for the subsequent course of Islamicate civilization on its intellectual
side.

In any case, the region from Nile to Oxus did remain distinct, culturally,
from the European region, where Hellenic culture — in Greek or in Latin, and

120 THE ISLAMIC INFUSION

even when Christianized — became the basis of all high-cultural tradition, such
as there was. The lands from Nile to Oxus were highly varied, and one cannot
say either of Egypt (where city life was especially Hellenized) or of the Oxus
basin (where Sanskritic influence was strong) a good deal that can be said of
the areas between. Yet the various lands in the region were closely interrelated;
to some degree they underwent a common destiny. I think this resulted at
least partly from the role of the mercantile classes in the regional life."

The position of the Nile-to-Oxus region in the Oikoumene

We have seen that the assertion of a regional identity between Nile and Oxus
was problematical, despite a certain amount of common history and culture.
It remained so even in Islamic times — though by then it was no longer the
Irano-Semitic traditions that were submerged by their neighbours, but rather
those traditions were overflowing throughout the hemisphere. It is not really
clear how our region came to have so anomalous a history. But we may guess
a part of it.

A distinctive high-cultural pattern in the region from Nile to Oxus was
surely encouraged by two geographical features of the region: its focal com-
mercial position in the Oikoumene and its relative aridity. These need not
have come into play in the origin of the distinctive patterns. Doubtless
the Irano-Semitic prophetic traditions originated in at most a few centres,
points where creative leadership was able to evoke unusually potent social
and intellectual norms, able to command wide human assent and imitation.
It is possible, for instance, that the uniquely creative tradition of the
Hebrew people was the outcome of a phenomenon rare in all Oikoumenic his-
tory: an enduringly successful peasant revolt (like those which launched the

11 It will be seen that I have been presupposing here a conception of civilization in
some ways more allied to that of Childe, say, or his disciple Turner, than to that of
Spengler or Toynbee. I see no necessary life-pattern in the formation of a civilization, and
I was willing to discuss the advent of a regional cultural integration among Semites and
Iranians without considering whether this was or was not a 'new civilization' : it was suffi-
cient to analyze it as a complex of cultural traditions in a new phase. Nevertheless, I feel
that practically all the analytic devices that Toynbee, in particular, introduces into his
system have their validity: a creative minority can indeed, given favourable circum-
stances, develop cultural patterns that will spread even at the expense of highly developed
rival patterns; and they can then evoke hostile reactions in those peoples among whom
they have been received. In particular, I see sufficient integrality in the culture of a
region to take seriously the phenomenon that both Spengler and Toynbee — building on
those historians who have contrasted the 'Greek' and the 'Semitic' spirit — have cited as
abnormal in this region: the seeming suppression by a Greek overlay of an indigenous
cultural life, which later reasserted itself (and most fully with Islam). But I do not think
that such a phenomenon can be reduced to general rules covering such cases on that level.
Certainly one cannot simply refer to latent spirits, either racial or cultural. One must
establish what it was, in a given case, that made an area receptive to alien patterns;
what gave it continuing interests calling for patterns alternative to the new dominant
ones; and what positive circumstances finally made possible the emergence of such alter-
natives.

THE WORLD BEFORE ISLAM 121

remarkable Druze and Swiss peoples). In that case, the covenant would re-
present the agreement of the rebels leagued against city domination, and
Joshua's campaigns would represent help brought to the peasants west of
Jordan by those initially successful in the east (perhaps, as in the Druze case,
under ideological leadership from Egypt). 12 In such revolutionary events
could have been revealed a deity not only ethical but supremely historical —
a trait uniquely characteristic of the god of the Hebrews. The implications of
such a conception of deity, even for urban conditions, would gradually be
unfolded in the dialogue of the tradition (as I have pointed out in the Prologue) .
Then such insights, once they had become established, could win assent far
beyond the original historical community. Yet whatever was the source of
the Hebrew tradition, in any case new viewpoints could become widely domi-
nant only in a general setting which would favour their survival as compared
with other viewpoints. In the case of the prophetic traditions, one can specu-
late that this setting was found among the Semitic and Iranian mercantile
classes more consistently than among either the peasants or their landlords;
then it gained its force from the ecology of the region.

The citied traditions of the region were first based on the agriculture of the
rich Mesopotamian alluvial plain, which required considerable investment in
controlled irrigation to be fully used; and to a lesser extent on Egypt (whose
irrigation required less management). For a long time, the wealth and hence
power of the regional empires was derived largely from the alluvial agricul-
ture; in the Mesopotamian plain, especially, investment became more and
more elaborate after the Axial Age, in part through the resources of cen-
tralized imperial power. Even elsewhere, agriculture was the major source of
revenue; associated with the court in profiting from it was a great class of
landholders who, as cavalry (sometimes followed by peasants from their lands
as infantry), provided much of the military manpower the ruling courts de-
pended on. But, except perhaps in exploiting the alluvial plain, there was an
early limit to any increase in the agricultural resources on which the land-
holders' position depended.

The region from Nile to Oxus forms the central portion of the great Arid
Zone stretching across the Afro-Eurasian landmass from the Sahara through
the Gobi deserts. Some large portions of it, as can be seen on the map of
physical conditions from Nile to Oxus, are pure desert — expanses of sand and
rock or of salt flats, where there is so little vegetation that neither humans nor
animals can live. Some few scattered districts, on the other hand, such as the
coastlands south of the Caspian, receive a great deal of rain, are naturally
heavily forested, and when cleared support a lush rainfall agriculture. But
most of the region is neither desert nor well watered but arid, in the sense

12 George E. Mendenhall, 'The Hebrew Conquest of Palestine', The Biblical Archaeo-
logist, 25 (1962), 66-87, se ts forth an alluring theory which has the virtue at least of
making intelligible the long-run story in ways that the supposition of a mass invasion
and settlement by pastoralists (itself improbable) never can.

THE WORLD BEFORE ISLAM 123

that water is the most decisive limiting factor in supporting life: the more
water there is, the more vegetation; the more vegetation, the more animal and
human life.

There are three types of terrain in an arid region: that where there is enough
scattered rainwater to produce sporadic vegetation for grazing, but where
settled agriculture in any one spot is impossible; that where rain in a given
spot is predictable enough to warrant planting crops — which, however, are al-
ways in danger of failing if the rain fails; that where water is available in con-
centrated amounts from some source other than local rain— as from geological
strata which trap water falling over a large area and then can be tapped by
wells, which serve to irrigate the fields. The latter sort of terrain is called an
oasis and may be large enough to include many villages and a city or two. The
same effect is produced when water is available from rivers fed by distant
rains: in this sense, the whole lower Nile valley can be called one large
oasis.

As compared with better-watered regions — including the greater part of all
three of the other focal regions of high culture in the Oikoumene — aridity re-
duced the resources and solidarity of those in any given area whose wealth was
in land. Within the cultivated terrain the yield was insecure in the rain areas;
and the irrigated areas, which required continuous investment, were widely
scattered for the most part, forming only a small part of the total land surface.
Then between settled agricultural areas or around them there was terrain good
only for herding grazing animals; and this terrain was often so extensive that
the herdsmen could not retain a village as base but had to take along their
families and develop a full-time social and cultural existence of their own, as
nomads in some degree. Such herdsmen yielded little or no revenue. Moreover,
they sometimes posed a source of social power which might rival that of the
agrarian gentry. (Of course, there have been many sorts and degrees of noma-
dism and of combinations of pastoralism with agriculture.) In any case, over
any given large territory human life was concentrated in favoured spots rather
than distributed at all evenly, so that it took more overall land space to sup-
port a given-population. Agricultural groups tended to be far apart, and those
who could be wealthy on an agrarian basis were relatively few in any given
place.

In contrast to the limits on expansion in agriculture, there seemed to be no
limits on potential expansion in commerce. Perhaps in part because it was the
point of origin of the Oikoumenic agricultural complex, but more enduringly
because of the Afro-Eurasian configuration of seas and mountains, the Fertile
Crescent and its surrounding lands offered juncture points for more long-dis-
tance trade routes than were concentrated in any other comparable region.
From a port (like Basrah) at the head of the Persian Gulf, for instance, com-
mercial relations were possible, directly or through neighbouring (Mediter-
ranean or Iranian) centres, with all the major regions of the Oikoumene; some-
thing impossible from a port in any of the other regions, as can be seen from

124 THE ISLAMIC INFUSION

the map of trade routes in the Afro-Eurasian Arid Zone. Nowhere did
merchants have more opportunity to become cosmopolitan in outlook — and
to gain wealth through entrepot trade whatever the local or even regional state
of prosperity. Over the millennia, as the Afro-Eurasian Oikoumene and the
area of citied commerce in it expanded, the long-distance trade throughout
the hemisphere became not only larger in sheer bulk but more varied. To the
steady accumulation of technical refinements in the lands of old civilization
was added the development of special resources in the newer areas — such as
the spices of the Malaysian archipelago; which areas then became themselves
commercial centres instead of mere way stations. Cumulatively, long-distance
trade bulked, millennium by millennium, larger in the economies of the var-
ious regions; the potential opportunities of merchants, particularly those at
the crossroads between Nile and Oxus, became correspondingly larger.

The fundamental economy of the region from Nile to Oxus remained agra-
rian-based and the absolute proportion of the wealth of the society going to
landholders remained larger than that going to merchants. But with the re-
latively low concentration of agrarian wealth and the relatively high concen-
tration of mercantile opportunities, it is possible that merchants in several
countries there (whether or not themselves engaged in distant trade) were able
to acquire a relatively larger proportion of the civil wealth than could mer-
chants in most agrarian-dominated areas; and over the region as a whole, in
any case, they were probably a bit less completely dependent on the local
landed gentry economically. Increasingly over the centuries there was oppor-
tunity for the. high culture focused on the market to become more autono-
mous and even more influential in the society as a whole. The form that such
influence took (as compared with mercantile influence elsewhere) was
governed, in turn, by the effects of the early advent of agriculture and of
urbanization in the region (and of its physical openness to widespread military
operations and imperial constructions) . An imperial tradition of long standing
accentuated the cosmopolitanism, but militated against its embodiment
in any sort of civic particularism. These tendencies were pushed yet
further in the Islamic period. (We will study them in more detail in Book
Three.)

It was doubtless a mercantile potentiality that, at one stage of Oikoumenic
development, made it possible for Greek culture to gain such a hold in the
region. It was doubtless the special conditions under which cosmopolitanism
developed between Nile and Oxus that produced a very different sort of mer-
cantile life there as the Oikoumenic complex matured and the region's role in
it became greater. In any case, in the field of religion, at least, the mercantile
classes were sometimes able to set the tone for whole peoples. From Nile to
Oxus, then, religious history especially took a distinctive turn after the Axial
Age, as we may see outlined in the chart showing the rise of confessional
religion (pp. 126-27).

THE WORLD BEFORE ISLAM 125

The confessional religious allegiances

Largely under the relative peace assured by the great empires that were built
following the Axial Age, and in part growing out of the great ideas as well as
the social ferment of that age, came the beginnings of a series of movements
which are spoken of as the great historical religions — the 'universal' or 'con-
fessional' religions. From an intellectual elite, the sense of challenge to the
individual conscience was generalized among the urban and finally (at least
superficially) even the rural masses. Sometimes the thoughts of great figures
such as the Buddha or Isaiah came to be directly accepted, in principle, as the
creative springs of the new traditions and as sources of inspiration among even
illiterates. Often new figures and new leadership intervened, launching re-
latively independent traditions. In either case, insights were woven into a
system, dramatized in exclusive myth and cult, and equipped with popular
organization and sanctions. The old local cults were subsumed under the new
allegiances, or were replaced.

By the early centuries of the Christian era were thus established, all across
the citied zone of the Afro-Eurasian Oikoumene, organized religious traditions
which, in contrast to most of the previous religious traditions, made not tribal
or civic but primarily personal demands. They looked to individual personal
adherence to ('confession' of) an explicit and often self-sufficient body of
moral and cosmological belief (and sometimes adherence to the lay community
formed of such believers); belief which was embodied in a corpus of sacred
scriptures, claiming universal validity for all men and promising a compre-
hensive solution of human problems in terms which involved a world beyond
death. Beyond this very general framework, the several allegiances contrasted
as greatly as possible. Especially those originating between Nile and Oxus, in
the prophetic traditions, contrasted in almost every possible way with those
originating in northern India, which ultimately shared with them the ad-
herence of most of the Oikoumene. But they all filled the same sort of social
role.

By the fourth and fifth centuries ce, these religious allegiances were not
only generally prevalent; the stronger of them, in their several areas, were
able to establish their representatives in some degree of political power. An
official form of Christianity achieved exclusive status in the Roman empire;
Zoroastrian Mazdeism gained a comparable status in the Sasanian (Persian)
empire, though had it to tolerate stronger rivals. Vaishnavism and Shaivism
(which together issued in modern Hinduism) vied for royal favours in the
Indie lands and southeast overseas. Buddhism vied with a Buddhist-in-
fluenced neo-Taoism for power in China. Even allegiances with fewer ad-
herents could dominate a court (like Jainism) or even control a kingdom now
and then, as did Rabbinical Judaism (at this time a proselytizing body like
the others) and Manicheanism. The origin of the latter illustrates the general
mood. It was founded in the third century in the Iraq, not to win over pagans

126 THE ISLAMIC INFUSION

The Development of Confessional Religion in the
Irano-Mediterranean Region, c. 650 BC-632 ce

THE AXIAL AGE AND FOLLOWING CENTURIES

Throughout the citied area of the Afro-Eurasian landmass arise
movements of independent thought, forming classical literatures,
which become the norm in culturally unified regional empires, in
which then spread the confessional religions (characterized by an
expectation of individual adherence on the basis of scripture)

c. 650-550 bc Zarathushtra (Zoroaster) and Jeremiah and others, in Iran and
Syria, as major prophetic figures demand ethical confrontation
of cosmos and history, while Thales and Pythagoras and others
in Greek Anatolia and Italy as pioneering philosophers pursue
rational investigation of human and cosmic nature; both
prophets and philosophers figuring as critics and reformers of
established nature cults

538-331 bc The Achaemenid Persian empire provides a single tolerant and
prosperous sovereignty from the Aegean to the Indus;
Zoroastrian tradition penetrates the priestly caste of Magi

433 bc Nehemiah restores Jewish worship at Jerusalem on a prophetic

basis, launching the Jewish community as a people founded on
faith in scripture

399 bc Socrates dies martyr to philosophy at Athens, becoming hero of

Greek humanistic idealism in its many schools

333-328 b c Alexander establishes Greek supremacy in the former

Achaemenid lands, initiating a long confrontation of the
Irano-Semitic prophetic with the Hellenic philosophic traditions

c. 200 B c- The Mediterranean basin is dominated and then ruled by the
200 c e philhellenic Romans, under whom Hellenistic municipal culture

is standardized, while the philhellenic Parthians dominate the
Iranian highlands and the Mesopotamian plain; common cults
offering personal salvation spread in all these areas

30 ce The Christian community is founded in Syria, universalizing the

appeal of the Jewish divinity

THE SASANIAN EMPIRE TO THE DEATH OF MUHAMMAD

Representatives of the confessional religions, from ocean to
ocean, achieve social authority and power, rivalling each other
for exclusive positions

226-642 The Sdsdnian empire replaces the Parthians in Iran and the

Mesopotamian plain, fosters urban prosperity with relative
centralization

THE WORLD BEFORE ISLAM 127

273 Mdnt dies, founder of the otherworldly Manichean faith, and

friend of the Sasanl emperor

285 After crises (235-268) in which the Roman city loses its

Mediterranean power, the Roman empire is bureaucratically
reorganized with its capital at the Thracian straits (from 330, at
Constantinople) ; Christianity persecuted as anti-social; rival cults
encouraged

275-292 In the reign of Bahrain II, at latest, Zoroastrian Mazdeism is

given an official central organization in the Sasanian empire,
and is allowed to persecute dissenters

324-337 In the reign of Constantine I, Christianity gains an official

position in the reorganized Roman empire, and subsequently
becomes legally enforced

485-531 In the reign of Qubad, Zoroastrianism and the Sasanian

aristocracy are torn by Mazdak's attempted egalitarian reform

c. 525 Christian Abyssinians occupy the Yemen, in alliance with

Romans, ending Jewish kingdom (which had persecuted
Christians)

527-565 In the reign of Justinian, Roman power and cultural

magnificence reach a peak, while the last Pagan school is closed
(529) and Christian orthodoxy is enforced

531-579 In the reign of Nushirvan, Sasanian power and cultural-

magnificence reach a peak, with the crown triumphing over the
nobility; heresies against Zoroastrian orthodoxy are stamped out

c, 550 Final break of the Ma'rib dam in the Yemen, symbolizing the

decline of the south Arabian agricultural society and the
predominance of pagan Bedouin patterns in the Arabian
peninsula

603-628 Last great war between the Roman and Sdsdnian empires, in

which the forces of both are badly depleted, but political status
quo is restored. Restoration of the True Cross to Jerusalem (629)
symbolizes triumph of Christian over Zoroastrian empire — and
over Jews and heretics

622-632 Muhammad, an Arab of Mecca, sets up a religiously organized

society in Medina, and expands it over much of the Arabian
peninsula to march with and even locally replace Sasanian and
Roman power

128 THE ISLAMIC INFUSION

(that is, adherents of the older types of religion) but to restore the pure truth
which was felt to have been already corrupted in all the other popular con-
fessional religious bodies. Almost everywhere, alongside the dominant reli-
gious bodies, there were minority groups ready to challenge the established
group if given the opportunity. Forms of Buddhism and of Christianity were
especially widespread. Taken together, these religious allegiances, extremely
varied as they were in their approaches, achieved in common one grand re-
sult: they eliminated (or took over and transformed) the old tribal and civic
cults, replacing them for public purposes with their own rites; and they
accustomed the people of most of the Oikoumenic citied zone to expect every
serious individual to acknowledge at least some sort of life-orientational
tradition of universal claims as ultimate authority in his life. In the western
parts of the zone, they even accustomed people to expect some such religious
allegiance to be not merely patronized but enforced officially by governments.
The Irano-Semitic religious traditions, which generally prevailed in the
western part of the Oikoumene, were of two families. What we may call the
'Abrahamic' religious communities, chiefly Jewish or Christian in various
forms, could be traced back to the tradition of the Hebrew prophets; they
generally recognized the act of faith of Abraham as their point of origin or as
their classic model. Among communities of Magian-Mazdean affiliations, the
most important was that of the Zoroastrian Mazdeans, who worshipped the
good Creator-god as Ahura Mazdah; Zoroaster was their great prophet. 13 The
Abrahamic traditions were expressed primarily in Semitic tongues and
flourished especially in the Fertile Crescent; those of Mazdean affiliations,
primarily in Iranian tongues and on the Iranian highlands. (Cf. the map of
language and religion.) But despite quite separate origins in Axial times, even
then the two prophetic traditions had begun to merge; by post- Axial times,
the two sets of traditions cannot be fully marked off from each other. The
traditions influenced one another and even converged; some groups cannot
clearly be placed in one family or the other. A different criterion is probably
more important for distinguishing among the Irano-Semitic confessional
religions, a criterion which only partly coincides with the distinction between
Abrahamic and Mazdean: their degree of populism.

The role of populism in the Irano-Semitic monotheisms

Some of the religious traditions emphasized more than did others what we
may call 'populism': that is, concern for the ordinary moral needs and cap-

IJ It is conventional to refer to this family of traditions as 'Iranian', but .such a term
carries unwarranted ethnic implications (and has misled some into supposing that other
traditions, such as Islam, have not really belonged in Iran) and is not exact anyway. A
reference to Magian-Mazdean affiliations covers all the traditions relevant here. It is be-
coming common to refer to those who applied to themselves the term Mazdayasnian, as
'Mazdean' (for brevity) rather than 'Zoroastrian'; accordingly, I reserve the latter term
for such aspects of the tradition as can be identified with Zarathushtra.

130 THE ISLAMIC INFUSION

abilities of the common people, as contrasted to the highly privileged classes;
or at least for what the religious specialists deemed appropriate to the common
people. All the confessional religious traditions may be called somewhat popu-
listic in that they tended to cast their doctrines and their moral standards into
forms intelligible to the ordinary person. But among some of the Irano-
Semitic religious communities, populist values were stressed even to the con-
demnation of any other values. I suggest that this emphasis seems to have
answered, in part, to the sense of propriety that prevailed among mercantile
classes when they distinguished themselves from the more aristocratic tastes
of the landed families. Merchants, preoccupied with their steady daily tasks,
necessarily felt themselves to be ordinary folk as compared to the court and
the landed aristocrats who had the wealth and leisure to set the cultural tone
to which all other classes were tempted to aspire. An assertion of their own
dignity must mean a certain rejection of that aristocratic culture that lured
or mocked them.

This is not to say that mercantile classes must be, or were then, always
populistic in orientation; but that populism was specially appropriate to mer-
chants when they were not the highest class, and yet were sufficiently autono-
mous to form their own standards. The sorts of individual whose imaginative
responses looked to the moral needs of common people had a better chance,
in a mercantile context, that their concerns would be reinforced by a corres-
ponding response in others. In the dialogue that formed the traditions, their
interpretations proved not only abstractly attractive but also relevant to the
material interests shared by all; hence while every moral view might gain a
hearing exceptionally according to temperament, this one would gain a hear-
ing from all, and come to dominate the common expectations cultivated in the
tradition.

Populism, at any rate, was strongest in the traditions that predominated in
the Semitic Fertile Crescent, where the long-distance trade routes were con-
centrated more intensely than in the Iranian highlands, and presumably the
relative cultural autonomy of the market was strongest. Accordingly, popu-
lism was strongest among the Abrahamic traditions, carried chiefly in the
Semitic languages, as the chart of cultural and religious orientations points out.
But the populistic emphasis could appear in either the Abrahamic or the
Mazdean family of religious traditions.

The overall outlook of both Abrahamic and Mazdean traditions may be
summed up as looking to justice in history through community. All the pro-
phets had stressed just action as the highest religious activity. In contrast to
the traditions of Indie origin especially, preoccupied with individual self-
awareness, the Irano-Semitic traditions (populistic or not) centred attention
on problems of interpersonal justice. Such an emphasis was crystallized in the
central doctrine of the Last Judgment at which every individual would be for-
ever rewarded in Paradise or punished in Hell according as he had done good
or evil in his lifetime. In contrast again to the Indie traditions in which right

3

o

13

Q

o

a

s

o

1

3 3

d 5 5
« g

< o
o
«
o

55

<

B

•3 £

d
•-3

o
w

55
<!

55
«!

S

3

o

£

s

£

3?

<!

«

0>

W

!>

to

2

3

<u

o

.d

R

+j

T3

a

a;

H
H

s^-S * *

^

<

h

l

o

3

§

60

a

d

u

O

o

O

t*

•a

tn

d

E

rt

<

c

.a

H
55

en

H

a

o

_o

CO

-t->

o

M
i-I

O

Cfl

H £

3

<* £

o

■a

'-3

o a

«

P5

T3

S o

C

S-fl

1

en

a
o

H ^

d
o

■-g

T3

. d

T3

2

o

So

<U d

ja ai

* 60

en ^

M to

El

CD

d

ni
S

'3

bo

d

„-8

m « S a

« 5
S-2^

o

i

d

+>

<u

M

13

fl

d

<D

<a

is

-p

en

«

d

to

d
o

• t-4

60

a

o
3

■3

"3

■a

T3

"tS

1H

d

ni

43

■s

T3

d

<D

en
O

«)

a

do

m

f— t

ni

>

2

60

d

|S)

S

£

cd

"3

*o

3

>J

«

«

u

132 THE ISLAMIC INFUSION

and wrong were to be balanced out in an indefinitely continued series of
reincarnations, in the Irano-Semitic traditions each person had one lifetime,
one period of responsible moral choice, which was irrevocably decisive. One's
moral decisions now determined one's eternal fate without appeal.

On the social level, such an outlook made for a strongly positive evaluation
of history. As in the individual life, so in the life of communities, what hap-
pened was irreversible and determined all future history one way or the other.
There could be no question of an infinite round of cycles in which gods and
humans were equally caught up. The way of the world in general and the
course of history in particular were governed by the purposes of one supreme
God, who expressed Himself in personal will and judgment, over and above
the regularities and recurrences of nature. The course of events itself was not
a matter of impersonal destiny but stood under His judgment: He was pro-
tagonist in a cosmic drama with a beginning and an end, in which the just
cause finally would triumph. In the Mazdean tradition, He was temporarily
limited by a counter-creator, author of evil; in the Abrahamic traditions, by a
less exalted enemy, the Devil; but these figures did not share His ultimate
cosmic status and were not, in any case, to be worshipped. Hence these faiths
can all be called 'monotheistic', as calling for the worship of a single dominant
figure. Any lesser cults that might be tolerated, as directed to angels or saints,
remained (in principle) subordinated to the one primary cult.

One responsible lifetime; one transcendent God — and one righteous com-
munity. To assert the priority of the moral universe over the natural called
for all three. With the cause of justice would triumph also the people which
had identified itself with that cause. What might seem like a tribal and
parochial heritage from pre-Axial times was transformed, in the monotheistic
traditions, into an assertion of social and hence historical responsibility. The
Jews were to bear witness, among the peoples, to the lordship of God; who-
ever did not join in the common witness but stood alone not only forfeited the
moral guidance the community could give him but was directly unfaithful to
the covenant made with God. The people of Ahura Mazdah were to help Him
clear the world of evil; who stood aside was a traitor. The cults of the old gods
of nature were superseded and even excluded by those of the new ethical and
historical God. The difference was expressed in a tendency for the new cults to
use less palpable symbols of the Transcendent, such as fire, rather than the
old fleshly images which answered well enough to gods of the visible world;
for the true symbol and expression of the divine presence was the community
itself. Monotheism might be defined by the worship of one God; it became
effective in history when the worshippers formed one people.

But this overall monotheistic scheme could be interpreted in a spirit more
populistic or less so. The ideal of justice, to begin with, could be interpreted in
harmony with an aristocratic social order. For the official Mazdeism of the
Sasanian empire, agriculture was the noblest of ordinary occupations; but it
was only the landed gentry, who depended on it financially, that formed the

THE WORLD BEFORE ISLAM I33

best of social classes, as its military defenders. The peasant was, in principle,
more greatly respected than the merchant or craftsman, but he was not ex-
pected to prove noble nor yet to understand the subtleties of religion, which
were reserved to a still higher body of hereditary and aristocratic priests. In
the cosmic drama, the priests, together with the landed gentry as a military
force, had the major part to play.

Yet Max Weber is surely justified in linking especially to merchants the
type of religion which stresses the ethical demands of deity more than its role
in ensuring the round of the agricultural seasons. 14 The demand of the pro-
phets for just dealing, originally perhaps of agrarian inspiration, could be
readily interpreted in terms of trade and contracts; so interpreted, it was more
immediately relevant to the market than any cult of nature deities invoked
to assure fertility and the recurrence of the immemorial natural order. The
monotheistic traits of the Mazdean tradition were always threatened by com-
promise on the top social levels: Sa.sa.nian state religion was never very pure.
It was when shaped by the populism of mercantile elements that what was
distinctive in the overall scheme of Irano-Semitic confessional religion stood
out most sharply against the general background of agrarian-based culture.
Nurtured in urban life, the monotheistic emphasis on an ethical God was
accentuated into an interpersonal moralism in maximum contrast to the
sympathetic ritual and even magic retained in the cult of the gods of nature.

In Judaism and related traditions, in which merchants held a much higher
position than in the Mazdean tradition, social justice accordingly took on im-
plications more challenging to the established order. It tended to be, above all,
egalitarian justice, the justice of the market, with every person equal before
the law. To be sure, marked equality, given the initial inequalities in human
nature, tends to issue in its own sort of social inequalities. The demand for a
more radical egalitarian justice, for effective equality among social classes,
was not always pressed very far by merchants who found themselves forming,
after all, a privileged class themselves. Still, several of the Abrahamic tradi-
tions (and even some religious traditions of quite different backgrounds else-
where in the Oikoumene) reflected initially an active sense of the equal dignity
and ultimate rights of the less privileged classes in society. Over the centuries,
within each Irano-Semitic religion, tradition tended to develop again and again
reformed versions in which such egalitarian justice was commonly stressed
and some degree of practical implications drawn from it. On the whole,
however, as they became established the religious leaders found ways to
justify accepting the social order for the present, with at best secondary modi-

H Max Weber's analyses are — in the nature of his purpose — schematic; even so, they
offer suggestions for possibilities to be alert for rather than universal rules governing
what can and what cannot happen. However unhistorical his discussions may be,
they offer an invaluable starting point for any attempt at analysis of the interrelations of
various social phenomena in particular cases. It is unfortunate that he said so little about
Islam. His 'Religions-sociologie', in Wirtschaft und Gesellschaft (Tubingen, 1921-22)
has been translated as The Sociology of Religion (London, 1965).

134 THE ISLAMIC INFUSION

fications. Any guilt felt for present unjustifiable privilege and good fortune
was to be assuaged by charitable actions and benevolent deeds; while the
social injustice itself was found to be a transient matter, counterbalanced, or
even eventually eliminated, in the cosmic order as a whole.

The populistic temper

The populistic spirit of the market was expressed more effectively in develop-
ing the prophetic ethical bent into a full-scale personal moralism: that is,
insistence on justice and equality in interpersonal relations even at the ex-
pense of any other sort of cultural value, such as those more imaginative
values which exceptional individuals or even whole privileged classes are
sometimes pleased to foster. The ordinary individual must be an honest man,
whose duty is to do what is useful rather than what is decorative, what helps
himself and others in the tasks of daily living rather than what embellishes
that living or even interrupts it. Aristocrats might indeed have the leisure to
exalt the creative, the adventurous, the tasteful; to value exceptional achieve-
ment in activities demanding special skill and talent. It was appropriate for
the aristocrat to aim at personal distinction in himself and to prize it in others,
including the learned among the priests. As to the merchant, he too was no
unsophisticated peasant; he had a high culture of his own. Yet he could not
normally look to the personal distinction which could come with a socially
more privileged position, nor even to the luxurious display which lavishly
patronized the arts. His culture must express itself through perfection in the
common duties.

The high culture of the market, therefore, emphasized more moderate vir-
tues: not personal courage or political or artistic virtuosity, but the 'bourgeois
virtues', thrift, sobriety, and, above all, respect for law and order. What would
require special talents and luxurious expense — the science presupposed in al-
chemy and in astrology, the art used in monumental sculpture and painting,
the extravagance displayed in silk and gold — was already suspect as inacces-
sible to the ordinary man. Carried into the realm of religion, this attitude rein-
forced the specifically spiritual objections that also arose to some of these
things because of their association with the old nature gods. In general, such
an outlook made for neglect of grand aesthetic ceremonial in favour of the
moralism encouraged in, say, the Talmud. At most, by way of special distinc-
tion, what the populistic temper could admire was an ascetic virtuosity, re-
quiring no lush resources beyond a personal self-discipline.

Just as for the populistic temper the monotheistic ethical emphasis led to
egalitarian moralism, maximally uncompromising in its contrast to nature
cults, so the monotheistic cosmic drama was concretized into immediate
human history, for the course of which each individual could be responsible.
The sense of cosmic drama could become still more definitive if set loose from
any analogy which tied it to the dramatic sequence of the seasonal year, and

THE WORLD BEFORE ISLAM 135

therefore to the natural order in which the landed aristocrat triumphed; it
could be interpreted entirely in terms of will and action, of bargain and ful-
fillment of the bargain. In this perspective, what mattered in the historical
drama was the action and fate of the religious body as such, as a group of re-
sponsible individuals. Such a religious group was made up of , and should even —
if possible — be led by common people, as the world reckons. It must be egali-
tarian; its adherents were essentially equal in status. Among the Jews, even
priesthood was in doubt: whoever was most learned and pious was rabbi, a
strictly lay leader. History required the human involvement of all. Finally, a
populistic temper could heighten the claims of the one religious community.
The aristocrat found his dignity through his place in the natural order of
society, where the common person counted for little. But in the community
of faith, the aristocrat's dignity need not matter; here the common person
could count, in principle, as much as any other. Since the decisive historical
responsibility was undertaken by the religious community as a body, all
human beings were called on individually to support it. The individual could
do nothing worthwhile outside it. Such a body was not only one of a kind, but
total in its claims. Ideally, every individual ought to live his spiritual life
entirely within its norms. With populism, then, the exclusive demands of the
religious body, too, became more comprehensive, universal, and uncompro-
mising. There was one true faith and all else was false.

Nowhere was the tendency for the lay population to be partitioned out
among the confessional religious bodies, as belonging exclusively and de-
cisively to one or the other of them, stronger than in the region from Nile to
Oxus. Even the more aristocratic-minded of the monotheistic traditions fell
into this tendency (perhaps by reaction to the others), if less rigidly than the
more populistic ones. Farther east, the religious specialists, even at the height
of their power, were never able to persuade everyone to declare an allegiance
to one system and only one. Though the more pious usually became convinced
that truth was entirely, or at least most perfectly, expressed in one or another
of the religious traditions, many of the ordinary population seem always to
have been inclined to respect equally the representatives of all the more popu-
lar traditions. In the European region, the religious traditions which first
introduced the new religious pattern were likewise usually tolerant of multiple
adherence; in any case, a person remained first a Roman and was only second-
arily an adherent of one of the new cults. Even when a single religious tradi-
tion — one of Semitic origin, to be sure — did win official and exclusive status,
it soon became so integral a part of the general culture that to be a Roman
and to be a Christian of the official church became almost equivalent. Be-
tween Nile and Oxus, on the other hand, the rise of the confessional allegiances
soon meant the organization of the whole population into many mutually
exclusive rival religious bodies; that is, into communities which were religious
rather than primarily territorial. It was as socially unthinkable to be associ-
ated with two or more such communities as to be associated with none.

I36 THE ISLAMIC INFUSION

Not everyone could be forced into the standard recognized communities. In
the environment of the prophetic monotheistic tradition, and stimulated by
Hellenic thought and notably by Plato's sense of philosophic wisdom, arose a
diverse movement, the most distinctive tradition in which is called 'Gnosti-
cism', because, it sought to liberate the soul from the darkness and false-
hood of material body through cosmic illuminative knowledge, gnosis. Gnostics
and those spiritually kin to them assumed that the universe was in all details,
as well as in its whole, humanly meaningful (whether positively or negatively),
and not merely an aribitrary creation of God for His own unfathomable pur-
poses; but the meaning lay not in the apparent structure of objects and
sequence of events but in a hidden truth that lay behind them, of which they
were the traces or perhaps the symbols. To fathom and realize the hidden
meaning of the universe, then, was the true calling of human beings ; this would
occur in a process of purification and enlightenment of the individual soul,
which was itself a part of a cosmic process of enlightenment in which the
whole universe was moving toward a truer condition, when its hidden meaning
would become manifest. Accordingly, the external conditions of the world,
including the popularly received religious traditions, were of little significance
save as they pointed the seeking soul toward the inward truth that lay be-
hind them. It was individual enlightenment that mattered; and this was
typically to be achieved not so much through a community allegiance as
through discipleship to a wiser individual who had already achieved enlighten-
ment, and whose enlightenment in turn came from discipleship to his pre-
decessor — and so on backward in a chain of discipleship to one who had re-
ceived direct revelation. For all that, the several movements of this type
ended by forming little religious communities or sects of their own.

The Gnostic movement and its spiritual kin were especially prevalent (among
a seeking section of the population) in the region from Nile to Oxus, in Egypt
and the Semitic lands and probably in Iran; but much of their writing there
was done in Greek and it was very influential also in the Greek and even Latin
European lands. It was in the early centuries after the Axial Age that these
several elite religious traditions took form, and many Gnostic sects were
associated with one of the major popular religious traditions, notably Christ-
ianity. But the most popular of the movements of this family (though it was
not a Gnostic sect in the exact sense) formed a fully independent community:
Manicheanism, which we shall meet again. All these movements shared a
number of common traits, despite their diversity, which allow them to be
characterized in common in relation to the best known of them, the Gnostic
(though indeed some movements which in this perspective can be associated
with Gnosticism had, in other respects, more in common with some other
group of tradition^ than with the Gnostic traditions). Or perhaps we should
say that tendencies of a Gnostic type, together with terminology typical of
the movement, were very widespread, so that traces can be found in almost
all religious movements of the time in the Irano-Semitic lands.

THE WORLD BEFORE ISLAM I37

Whether in esoteric or, more usually, in exoteric form, it was in terms of
religious communities that the Nile-to-Oxus region was maintained as one of
the four core areas of high culture after the Axial Age. When the Semitic and
Iranian traditions were overlaid by the Hellenic traditions, they received
little support from common administrative or even commercial continuity.
Rather than in a common classical language, the persistence of an indepen-
dent high-cultural orientation was expressed, above all, in the various religious
communities, comprehensive as they were and demanding exclusive total
loyalty. Eventually, each major religious community tended to cultivate its
own more or less localized literary language, usually some form of Aramaic or
Iranian. In each of these languages the traditions going back to the prophets
of the Axial Age were more or less independently developed. Commonly,
selected writings from those prophets were retained, at least in translation;
sometimes the tradition was reformulated entirely. Greek long remained the
most widespread common language of high culture in most of the region be-
tween Nile and Oxus, but — outside of certain explicitly Hellenic cities and (in
the Roman provinces) certain elements in the upper classes — it was increas-
ingly limited to circles concerned with philosophy and science; and even these
subjects were increasingly cultivated in the Aramaic and Iranian languages of
the various religious communities.

The confessional empires

The cultural independence of the Iranian and Semitic peoples from the Hel-
lenic cultural sphere was at best imperfectly expressed, on the political side,
in the Parthian Arsacid empire of Iran and the Iraq, which had succeeded
outright Hellenic (Seleucid) rule except where Rome took over (in Syria). But
in the third century ce, with the increasing predominance of the confessional
religions, both Roman and Parthian empires were transformed in such a way
as to give the Irano-Semitic traditions more complete expression (though not
necessarily yet in their most populistic form).

In the Mediterranean, the old empire of the Romans was replaced, after a
period of near anarchy, by a structure based more equally on all the Mediter-
ranean peoples, with a capital, Constantinople, nearer the centre of economic
life and of the old Hellenic culture than Rome had been. It was then that the
reconstructed Roman empire adopted a new confessional religion, Christianity.
It was this altered Roman empire into which, later, Islam came.

Greek was the empire's chief cultural language, though the Latin of its
western provinces was long maintained throughout the empire as the lan-
guage of law and in some other spheres. The classics looked to by both Greek
and Latin provinces of the empire (including, of course, large elements in Syria
and Egypt) were the Greek masters, from Homer to Aristotle; and the political
and social ideals were still, even when in fact the emperors were almost abso-
lute masters, traceable to the ideals of liberty and civic virtue of the ancient

138 THE ISLAMIC INFUSION

Greek polis. Since the beginning of the fourth century, however, the empire
had been officially Christian. Though the Christian New Testament was writ-
ten in Greek, Christianity had its origin in the Semitic Fertile Crescent. (Only
gradually did the Christian bodies of Syria and Egypt form themselves into
communities distinct from the official church of the more Hellenic and Latin
parts.) The spirit of Christianity — and the massive institution of the church —
had long competed strenuously against the main classical Greek tradition for
the allegiance of the ruling classes. Hence the Christian triumph marked a new
era. Hellenistic art, philosophy, even science and law were remoulded under
Christian influence in the Roman provinces as well as farther east. It was in
this form that they reached the Muslims.

By the time of Muhammad the most striking feature of the Roman empire
was its support of the Christian church over a vast area, even though in a
form that had become unpopular in Syria and Egypt (and among some
elements even in the Maghrib) . In his time, the Roman empire had lost a good
many provinces in the west, holding there only southern Spain, the Maghrib
(North Africa), parts of Italy (including Rome), Dalmatia, and islands like
Sardinia and Sicily; but it still controlled the seas; and in the east its hold on
the Balkan and Anatolian peninsulas, Syria, and Egypt was practically in-
tact. 15

Meanwhile, in the same third century which saw the reconstruction of the
Roman empire in the Mediterranean lands, the Parthian dynasty was replaced
in the Iraq and Iran by the militantly Iranian dynasty of the Sasanian
Persians. 16 As becomes clear on the chart of the origins of the Islamicate culture,
the Sasanian empire was the great rival of the later Roman empire to
the east, and the chief predecessor of the caliphal state. The Sasanian capital
was Ctesiphon on the Tigris, near what was to be the Muslim Baghdad and
about forty miles from what had been the ancient Babylon. The Sasanian
Persian nobility looked back to the ancient Achaemenid Persian empire as its
model. The Sasanian territories were not so broad as the Achaemenid had
been, particularly in the west where, far from extending into Europe, they
did not include even Syria and Egypt. These latter two provinces continued —
under protest, to be sure — to be ruled by the Roman empire, and hence by the

15 It is important to remember that the Roman empire which the Muslims confronted
was not an 'Eastern Roman Empire' but the continuation, though weakened, of the
whole empire in the Mediterranean as a unity (compare the section on world-historical
usage in the Introduction) .

16 The Sasanian ruling class is often referred to as 'Persian', as is also the empire itself,
because the origin of the dynasty was in Fars, ancient Greek Persis in southwest Iran, as
was that of the Achaemenids; and the official language was Parsik, 'Middle Persian', mis-
called 'Pahlavl'. This usage of the term 'Persian' to refer to a connection with Fars, in
whatever age, is related to but distinct from a usage of the term commoner in this work:
to refer to the Farsl, 'New Persian' language and the extensive populations using it,
whether or not under a dynasty from Fars. Provided the term is used with care, the two
usages need not conflict: the first is for pre-Islamic times, the second for Islamic times.
When it is carelessly used, however, considerable confusion can result. (Cf. Note 3
in the Safavi chapter below, Book V, chap. 1.)

ft
o
l-l

s
W

a

c

3

_o

>,

+j

■a

■a

3

d

d

u

ffi

•4-*

<u

TJ

+->

o

D

rt

■a

■w

T3

c

en

3

3

45

w

5

en

.5

o-o

01

rt

*£h

N

A

U

-t-> •£

^£

U O

H O H

O *■* 2
U5 -B J3

I ^ +3
So , 3 °

. 3

«!

3 5

o rt

(-4 3
^ tn

©■p S;

S3 ftc3

"> o .2

d R +j

'S ri 2 3
<d e +3 Cu

■a'S-o u

a 5 « k

d n p-

SS a £

w ti « ij

§3

Bg'3
C d t-<

a d d

° t!-

.2 -a n

60 w H
* « 3

13
<d

.a
3

0)
ID U

Ssa-8

.2 «
SjP4
o o <u

sfi

d en c
'■CSS
.2 5
.3,3 bo

s ° a

i-l d «>

3 u

3 d «j

CD O M

■Ho «
ft+i

oo-n 2
»S o

I >d *i

OD'd iw

B.S
•rt 60

•a
c

• d

"go

C5 "+^
-* d o
jj E +j
fe O d

° -h

OB,

d 6

6

o

d

a

c

>.

o

■a

'S

.^

3

3

a>

If)

(«

u

4i

S3

T3

00

C

to

3

0)
3
C

S S «

O d g
k. rt <i>

d ;S 3

S S 8
S « K

h6b

GO » S
r-(J3 O

(0V U

2 <D

s.h

ci

3 d"

lug

OTW o
«*«

n! O >d d)

d
<u.3

3i2"^

i§l

2 o ^

«

ci

I

.■ajB
mft ft

+* -*i

£ <D

s ^
^ 2*0

^^ c

« d u

p > ?
- 'S o

u 1) 4->

111
° B B

ss'3

5 a

cd <D ^

d . E
MOO

oa r r; •-■

CD +j 'O

o

te.s

§^

n-t *

•2 §

s ft

D ra'O

s o a

ft rj 3

d aj o
"fc Q w

CO uj d

-« i—l H
o r»<3

ho

Oi

5^

O aj
<* ft

^4 „

S Sfw

d 3 ^

.'-' 'o °

* « S

d a) £

I 3 a

odd

S.S(§

§SJ

IS

uu

§1

u d
O

&.

rt r" C jj

K u o

O O « +J

3 a 1-y «

+J ^ « c
--i^w o c d

„§3cd

us B S« j3

t> 3

S3-
a> g a) P.

I d> QJ to

o ftftn
oioo q
us ft.gcj

.2 "i

- S 2

tlfe en
d _ -t->

<u 53 ^

Bg-dg

m 3 C

* (u o 5

J.S aoi3

r> S

-S o
■3 n

140 THE ISLAMIC INFUSION

chief carriers of the Hellenistic traditions as modified by Christianity. Never-
theless, the greater half of the Fertile Crescent was Sasanian, as well as most of
the Iranian highlands; that is, the majority of the centres of Irano-Semitic
culture; while from time to time Sasanian power extended beyond the moun-
tains in the north and east, around the coasts of Arabia, and for a moment to
Syria, Egypt, and even Anatolia.

There were two chief languages of high culture in the Sasanian domains,
one each of the Iranian and Semitic families, as had been the case in the
Achaemenid empire and as was to be the case under Islam. Of the Iranian ton-
gues, Pahlavi (properly Parsik) — spoken in the highlands, especially Fars in
the southwest — was the official language of the court and of Mazdean religion.
It was in Pahlavi that translations from Indie literature made themselves most
felt. Of the Semitic tongues, a form of Aramaic called (eastern) 'Syriac', used
especially by the Nestorian Christians of the Iraq, was the chief carrier of the
Hellenistic traditions apart from Greek itself. Many works, especially on
natural science, were translated from Greek into Syriac; and some also into
Pahlavi. (The Nestorian church, opposed to the official Roman, Chalcedonian,
church, was commonly favoured by the Sasanian empire, among its Christian
subjects, as being in natural opposition to the Roman empire.)

But (in some contrast to the older Achaemenid — and especially to the sub-
sequent Islamic — situations) the two chief languages carried largely different
cultural traditions. Moreover, several other literary languages were also used
in the region as a whole, partly based on religious communities and partly on
local traditions, as in the Syr-Oxus basin to the northeast. In particular, other
forms of Aramaic were used by Jews and by other Christians, including those
across the boundary in the Roman empire who yet dissented from the estab-
lished church of the empire. (The dissenting church in Egypt used the local
Coptic tongue.) Sasanian Iraq was the centre of Jewish life throughout the
Oikoumene; there the Talmud was being compiled (in a form of Aramaic).
Jews formed a substantial part of the population of the cities and countryside
of the Iraq.

Though the Sasanian court looked to the ancient Achaemenid empire as a
model, the new empire was far from a restoration of the old, for the older
empire had naturally not embodied all the cultural forms that had later re-
sulted from the ferment of its own time, the Axial Age. Into the Sasanian
ideas of culture and the good society entered many traditions not only
from Achaemenid times but sometimes traceable back even to the ancient
Babylonians. The Sasanian hope f,or a universal absolute monarch, whose
disinterested power could curb all lesser ambitions and allow peace and justice
to the ordinary population, thus went back to Achaemenid experience. Yet
even the ideal of the grand monarchy was no repetition of the tolerant
Achaemenid overlordship. It had inherited the large-scale economic initiative
of the Hellenistic monarchies (the name of Alexander played as great a role in
the Iranian as in the Roman imagination); and its concern with central power

THE WORLD BEFORE ISLAM 141

reflected, perhaps, dissatisfaction with too loose a central control under the
later Parthians. Sasanian absolutism was a distinctively post-Axial political
order, typical (in essentials) of the imperial regimes of the time. Above all, it
was supported by an official confessional religious allegiance, which made for
cultural unity in the empire at least among the ruling classes. 17

About the time that Christianity was being established in the Roman em-
pire, Mazdeism was being given a corresponding position in the Sasanian realm
by early rulers of the dynasty. In this tradition, men revered Zoroaster as a
supreme prophet, and his poetic compositions, the Gathas, afforded perhaps
the most creative impulses taken up in it. Ahura Mazdah was worshipped as
lord of light and truth in the struggle of creation against darkness and false-
hood. At the same time, in coming to offer a comprehensive religious pattern,
the tradition had built upon much else as well. It centred on the learning and
the ritual offices of a hereditary priestly class, the Magi, guardians of the
temples in which a sacred fire was kept burning as a symbol of the light of
Ahura Mazdah, and in which the several angelic expressions of divine truth,
'Good Thought' and other guardian spirits, were duly invoked and honoured.
On the official level, the Magi had to tolerate quite a pantheon of old gods, in
fact. In the life of the more scrupulously pious, the faith expressed itself most
strikingly in a minute and comprehensive concern for purity — for physical
ritual purity, and for the purity of a truthful mind.

As we have noted, official Mazdeism was agrarian-oriented and aristocratic
in temper. Justice was to be found in a well-ordered hereditary class society, in
which each class was to have its own dignity and its own reward. The peasants
were honoured as performing the most essential tasks in the production
of basic needs — the cultivation of the sacred earth, to bring forth nourishment
for life. Yet they were justly to yield what of their product they did not need
for their own lives to the landholders, the government officials, and the Magi
themselves; the privileges of these latter were justified by their functions in
maintaining order, defending against alien predators, and mediating between
humans and the divine. Other privileged groups seem to have had less dignity
in the official system — the urban merchants and craftsmen, caterers to the
desires of the rulers and priests for cultivated luxuries. In theory, the more
frivolous arts of civilization (but not the priestly art or the art of government,
regarded as cosmically useful) were given little place, and their representatives
regarded almost as parasites. In practice, because a privileged position was

17 On the Sasanian empire — and generally on the Fertile Crescent and Iran in the thou-
sand years before Islam — we are more poorly informed than perhaps on any other major
lettered historical period. On the Mazdean religious tradition, see Jacques Duchesne-
Guillemin, La religion de I'Iran ancien (Paris, 1962), especially the bibliographical
discussion in the chapter called 'Histoire des etudes'. For a survey of the Sasanian
empire see Arthur Christensen, L'Iran sous Us Sasanides (2nd ed., Copenhagen, 1944),
somewhat uninspired but indispensable. For comments on making use of the work of
Zaehner and of Altheim, two of our most creative scholars in dealing with the period,
but whose works are subject to great caution, see the section on the history of Islamics
studies in the section on historical method in the Introduction.

142 THE ISLAMIC INFUSION

granted to some, those who created the luxuries of privilege were indirectly
justified also.

Populism in the Sdsdnian empire

The Parthian empire had been a relatively loose association of regional govern-
ments and semi-independent cities, among which the central dynasty had
failed to prevent frequent warring. Probably wealth was less concentrated
than in some other periods and we have little evidence of great monumental
cultural works. Nevertheless, it seems to have been a time of considerable
vigour, both economically and in arts and letters. The foundations then laid
come to spectacular fruition under the Sasanians.

In the Sa.sa.nian period, peace was usually maintained within the empire.
Monumental works of building and sculpture, technically masterful and beauti-
fully expressive of empire, bear witness to great concentration of wealth —
that is, great prosperity for the privileged (which might or might not be
accompanied by prosperity for the ordinary peasants, of course). Economic
development was twofold. On the one hand, agrarian investment reached its
peak in the Mesopotamian alluvial plain, with massively large-scale irrigation
and a steady increase in the agricultural population. On the other hand, mer-
cantile trade was extended and industry fostered; for instance, both the im-
port of silk from China and its working within the empire. Cities increased in
wealth and importance; the Sasanian monarchs were notable as founders of
cities and protectors of trade. 18

The mercantile development represented in part a response to the ever-
quickening pattern of trade throughout the Afro-Eurasian Oikoumene. Direct
trade by sea and land between China and the Indo-Mediterranean regions
opened up only at the end of the Axial Age, yet quickly became important
commercially and financially. Trade elsewhere in the Southern Seas {the seas
of the Indian Ocean and eastward) had likewise expanded, as had trade both
north of the Mediterranean in Europe and south across the Sahara. The
peoples from Nile to Oxus not only took full advantage of their crossroads.
They helped develop the new fields of trade. It was in this period that Greek,
premier commercial language at the end of the Axial Age in many parts, was
displaced in favour of elements from the Fertile Crescent. Along the central
Eurasian trade routes, it was Iranian and Semitic culture and religion that
came to dominate locally even more than Chinese or Indie. Along the west and
especially southwest coasts of India, it was in this period that Christian and
even Jewish populations — settled or converted from the Persian Gulf area —
began to become numerically important. In the west Mediterranean and its
hinterland, the same elements — 'Syrians' and Jews — became carriers of mer-

18 On the prosperity of Sasanian Iran and Iraq, see Thorkild Jacobsen and Robert M.
Adams, 'Salt and Silt in Ancient Mesopotamian Agriculture', Science, 128 (1958),
1251-58.

THE WORLD BEFORE ISLAM I43

cantile culture to the point that even under local Christian rule Judaism
proved to have great attractiveness for converts. Our region was becoming
the mercantile heartland of the Oikoumene.

It should perhaps not surprise us, then, that especially in the Fertile Cres-
cent the agrarian-based imperial powers should find themselves challenged by
populistic movements (sometimes, at least, of a mercantile cast) expressed in
the confessional communities. In the midst of all the prosperity, religious con-
flict became a major theme of social life. Not all dissident religious groups
were subversive. Of the numerous sects of Gnostic type that had arisen es-
pecially in Jewish and Christian contexts, at least some seem to have been
relatively populistic in temper, expressing clear social protest, but not rebel-
lious. The same may be said, it seems, of the Buddhist groups found in much
of the empire and in the Syr-Oxus basin (in fact, Buddhism seems in its own
way to have been as populistic as any of the other confessional traditions) . It
was early in the Sasanian period that the Manicheans attempted their vigorous
synthesis of all previous revelations in a system of Gnostic type, and spread their
gospel wherever the Iranian and Semitic merchants went. For a time, Mani-
cheans, and evidently representatives of other movements also, hoped to win
the Sasanian monarchy to their support. But after official Mazdeism came to
be clearly established, unrecognized communities were allotted an inferior
position or not tolerated at all; the Sasanian empire proved committed to the
agrarian gentry which provided it with its primary armed forces — heavy
cavalry. It became clear that a shift in the religious — and political — establish-
ment could be made only through an overthrow of Mazdean power.

There were evidently several movements which combined religious innova-
tion with social protest against the privileged classes. The most effective of
these was that of Mazdak: a leader who seems to have won the monarch him-
self to a programme of popular semi-egalitarian justice against landed pri-
vilege. Despite the relatively centralized and bureaucratic structure of the
Sasanian state, the old noble families had inherited a great deal of indepen-
dence from Parthian days; weaker kings were ruled by them, and stronger
kings had to carry on a constant struggle to assure the enforcement of central
policy. In the latter part of the fifth century, the monarch Qubad (Kavad)
was evidently persuaded for a time to try to undermine the nobles' growing
power by supporting Mazdak's movement to destroy or cripple the grander
forms of social privilege in the name of ascetic spiritual fraternity. Many
nobles lost much of their property (and even their superfluous wives), and
commoners were raised to high positions. For some years, the state was in
great turmoil.

In his later years, Qubad abandoned the movement, and his successor,
Khusraw Nushirvan, led a reaction. Mazdak and many of his followers were
massacred, the old official Mazdeism was restored, and the nobles given back
their privileges. But evidently their old position was not fully regained. The
nobles' power to disregard central authority was, at least temporarily, re-

144 THE ISLAMIC INFUSION

duced; Nushirvan seems to have been the strongest king of the dynasty. He
used the opportunity to reorganize the whole empire, setting taxation on a
more commercial basis; in particular, he seems to have increased state invest-
ment in the Iraq alluvial irrigation agriculture. It was probably in his time
that investment there reached its absolute peak.

Thus financially buttressed, Nushirvan was evidently in a position to develop
a stronger central army, supported by taxes rather than levies of the gentry.
We know as yet too little of the reign, but one source of recruitment for that
army was pastoral Arab tribesmen, who were independent of links to the
agrarian gentry, accustomed to fighting, and easily accessible from the Iraq.
The Arab pastoralist society was, of course, inherently fragmented, so that
reliance on Arab soldiery (even if still tribally organized) seemed to pose no
political threat comparable to that posed by the gentry. (Within a few years
after Nushirvan's death, however, the Arabs were interfering in the succes-
sion.) Thus the court and its military and civilian bureaucracy, with perhaps
the co-operation of some of the wealthier mercantile elements of the cities,
presumably gained a certain independence of the landed nobility, and perhaps
even outweighed it. Nushirvan was subsequently revered as the model king of
a model dynasty, the supreme exponent of royal justice. Muhammad was born
in his day, Muslims were proud to note. The centralizing, urbanizing ten-
dencies which had distinguished Sasanian times from the start came to a
climax under his rule.

Nushirvan's time is traditionally made also the peak of Sasanian literary
culture. Monumental history of ancestral legend and royal deeds paralleled
the monumental visual art. In a lighter vein, courtly literature seems to have
cultivated a wide-ranging curiosity; and Hellenistic culture was revived, not-
ably at the school of medicine and philosophy at Jundaysabur in Khuzistan
(the southeastern part of the Mesopotamian plain) ; some Greek philosophers
were attracted there for a time when Justinian closed the schools at Athens.
Its science and philosophy found favour even among some Mazdean priests,
who developed their theology in terms of it. But the most important source of
intellectual stimulus at this time was northern India. In the first centuries of
our era, there was a broad extension of Indian commerce and Indian ideas, to
about the same extent, though over a somewhat different area, as Hellenic
ideas had spread in the last centuries bc. It was this movement which had
carried Buddhism, at the hands of active missionaries, to wide popularity in
eastern Iran and central Eurasia before it spread throughout the Eurasian
Far East. (Whether, indeed, the early Indie development of monkish ways
accounts for the spread of various sorts of monasticism even in the Mediter-
ranean basin during these centuries is not fully clear.) Under the Sasanian
empire, and especially in its last century, Indie ideas in science and the arts
became popular. About this time Indians had made notable advances in
mathematics, especially over the Babylonian and Greek foundations they had
largely built on, as well as, it seems, in medicine; this seems to have been at

THE WORLD BEFORE ISLAM I45

least partially appreciated in Sasanian scholarly circles. Wider circles accepted
Indicism to the point of making popular lighter Indian literature in transla-
tion, and such a cerebral Indian game as chess.

In the early centuries of our era the civilized lands west of India were all
overshadowed culturally, if not ruled directly, by the Sasanian and Roman
empires. By and large, the effective area of the Sasanian empire tended to be
slightly smaller than that of the Roman, but they met as equals. Despite the
internal problems of both empires with dissident communities, they main-
tained impressive rival power structures, each claiming {in principle) universal
authority. The two empires struggled repeatedly with each other along their
frontier on the upper Euphrates, often giving only grudging attention to other
opponents; in the case of the Romans, to the Germanic and Slavic peoples
of inner Europe, in the case of the Sasanian Persians, to the Turkic states of
central Eurasia. In Muhammad's lifetime they waged the most destructive of
all the wars waged against one another in all their centuries of fighting, and at
his death both empires were financially and politically exhausted. 19

" Gustave von Grunebaum has saved me from some errors in this vast field, where I
am ill at ease (but he cannot be blamed for the persisting errors or biases, here or else-
where) .

Muhammad's Challenge, 470-624

The lands from Nile to Oxus would most certainly have changed somehow
in the eighth and ninth centuries, even without the intervention of Islam. The
Sasanian empire, perhaps under a new dynasty, might indeed have succeeded
in taking Syria and Egypt (if not the Maghrib) permanently from the Roman
empire and might have developed, in Syriac (Aramaic) and Pahlavi (Iranian),
a culture not unlike what developed in fact among the Muslims in Arabic and
Persian. Such a culture might have been carried fairly widely in the hemi-
sphere; for some elements of the Iranian and Semitic traditions were already
being carried into Europe and India. But it is hard to conceive of such a re-
newed Syriac-Pahlavi civilization as having developed all the homogeneity
and expansive vitality which were manifested under Islam. Arab tribes might
even have taken a hand in establishing the new dynasty which would have
succeeded to the Sasanian, yet by all analogies we must suppose that ordin-
arily they would have been rapidly assimilated to the more cultured settled
population, forgetting in time their remarkable but limited Bedouin poetry,
learning to speak some sort of Aramaic, and adopting one or another of the
existent forms of Christianity. One ingredient, the presence of Islam, would
seem to have made the vital difference, making possible a truly new civiliza-
tion, based on uniting the bulk of the population of the region into one reli-
gious community.

Among the most important elements in the background of the Islamicate
civilization, then, is the development of Islamic religion and of the community
which carried it. This was the work of a number of remarkable men, starting
with Muhammad himself. The community was built up first within the general
Arabian culture and then, after its sudden conquests, as a ruling community
scattered thinly throughout the Aramaic- and Iranian-speaking lands and far
beyond. Within the community there were sharp disagreements about what
its character was to be, and many struggles among the contending parties.
The Muslim religion and community that resulted from these struggles could
not have been foreseen by any human means; yet they bear the impress of the
vigorous minds and devoted spirits that went into their formation. 1

1 William McNeill, in his Rise of the West (University of Chicago Press, 1963), did me
the honour of referring to an early (and very incomplete) version of this work as an
important source for his own thinking about Islam and the Islamicate civilization. This
is flattering, for I regard McNeill's book as very important in the sense that it is the
first genuine world history ever to be published (the first to present the history of
citied peoples as a single overall historical complex, with primary attention to interrela-

146

MUHAMMAD S CHALLENGE I47

Bedouin-based culture in Muhammad's time

Between the Roman and Sasanian empires, and increasingly important in their
wars and their commerce, was the vast bloc of Bedouin Arabia. This was not
simply the Arabian peninsula as such. Bedouin Arabia was that area of the
peninsula in which the customs founded on camel-nomadism prevailed: pri-
marily the north, west, and centre. These were arid steppe lands, interspersed
with great reaches of rock or of sand, visited in winter and especially spring
with sporadic rains that awoke transient vegetation. The steppes were dotted
with oases, where the earth formations brought water fairly close to the sur-
face in sufficient amounts so that regular irrigation could maintain a more or
less extensive agriculture in limited and isolated areas. From the spotty sea-
sonal vegetation, and from springs and wells that could be dug even where no
oasis was possible, by keeping frequently on the move, herdsmen could main-
tain their animals, supplementing their milk and occasional flesh (and what
they could get by hunting) with the wheat and dates that agriculturists could
raise in the oases. The agriculturists, in turn, though hemmed in by the desert,
could get needed animals from the herdsmen and also any specialized products
that required bringing from a distance. But such a pattern of living pre-
supposed the domesticated camel, alone capable of forming the basis for large-
scale pastoral life in such deserts.

The camel does not seem to have been domesticated early, not before the
second millennium before our era. Only after long experience in breeding
camels for transport would some pastoral groups, relying less on sheep and
more on camels, have been able to move out into the more arid regions at a
distance from the settled areas of the Fertile Crescent, followed step by step,
presumably, by the necessary agriculturists in the oases. Independent
pastoralism may have begun not long after agriculture itself — that is, a life
based on herding in which the herdsmen are not members of a village
community, herding the village animals, but rather form permanent social
units of their own independent of any particular village. But camel nomadism
deep in the desert was a highly specialized form of life, presupposing special
technical and even social skills. We cannot surely identify camel nomads,
that is, Bedouin, until the beginning of our era. By then, the tradition was
well established, at least in the margins both of Syria and of the Yemen,
and it unfolded its consequences rapidly from that time on. 2

tions among the societies, and without unduly excessive attention to one society),
Unfortunately, I find myself occasionally disagreeing with McNeill (beyond this funda-
mental point) both as to basic theory and as to the interpretations of the several
civilizations; and particularly as to his interpretation of Islam and Islamdom. Many
of my points of disagreement both on world history generally and on Islam will become
obvious in the course of this work.

2 I owe to H. A. R. Gibb's lectures my attention's being drawn to the significance of
the chronology of camel pastoralism, as well as to numerous other points in this part of
the work. The chronology itself is still in doubt.

I48 THE ISLAMIC INFUSION

Once camel nomadism had developed, it carried potentialities of a major
social force. Camels allowed their herders greater mobility than other pastoral
animals, being able to endure longer than ordinary animals without food and
even water, and so to travel farther between watering places. But the wild ass
is almost as tough, and faster; the hunting people who tamed it could range
even more freely than the camel men. It was crucial that the camel was also a
great beast of burden: it was unrivalled, except by the elephant; and it
yielded good milk; through its various qualities it not only sustained its
owners but found a ready market. Such economic advantages enabled a
heavier concentration of people to live from camels than could live, say, from
wild asses; but with equal independence from agrarian controls. This gave the
Bedouin a potential predominance over not only the desert oases but even the
nearby reaches of the settled countries, allowing them not merely to trade
(which was essential) but, under favourable circumstances, to exact tribute.
If there were governments strong enough to refuse tribute, then the qualities
that Bedouin life developed in the men and their ready-made transport equip-
ment and their numbers made them welcome soldiers for those same govern-
ments — which could prove an alternative path to social power.

The Bedouin necessarily developed their own distinctive type of social or-
ganization. Originally adapted to herding, it could be maintained more or less
in other situations too. In fact, by the sixth century even parts of western
Arabia that had once been little more than extensions of the settled life of
Syria or the Yemen, living under kings, had been absorbed into the Bedouin
life; even people settled in the agricultural oases or in commercial towns tended
to be organized as 'settled Bedouin', to keep camels, and to think of them-
selves as if they were in principle pastoralists.

This sixth-century Bedouin and Bedouin-based Arabian culture differed
from that of the more agriculturally developed of the lands on the agrarianate
level in its presuppositions for historical action. Bedouin-based society pre-
supposed the wider agrarian-based society of which it was essentially an
extension; and hence looked to an agrarianate level of high or learned culture,
if to any. It did not escape the overall historical limitations imposed by the
fact that the bulk of the resources on which any large-scale historical endea-
vour must be based were limited by the agricultural resources on which wealth
and leisure were ultimately based. Yet within these limitations, Bedouin-
based culture posed a special variant case. With all pastoralists on roughly the
same economic level, there might be wealthier or more respected families, but
relative to more agrarian societies there was little class stratification and con-
centration of wealth — the herdsmen could not be so readily exploited as could
a peasantry — and hence also a lack of many aspects of learned high culture
within the Bedouin communities.

The relative equality among Bedouin was reinforced by their tribal organi-
zation, such as independent herdsmen have commonly developed; that is,
hereditary economic and social solidarity among smaller or larger groups of

MUHAMMAD S CHALLENGE 149

families, not based either on territorial proximity or on directly functional re-
lations, yet sharing a common responsibility in good and bad fortune. In this
way, families were associated in larger groups for general economic purposes,
and these in turn in still larger ones for political strength. Groups on every
level possessed internal autonomy, but were likely to be grouped with yet
others in still larger associations. (We call the larger — and more tenuous — of
these groups 'tribes'. Smaller groups are sometimes called 'clans'.) At every
level, these groups defined themselves in terms of a real or fictive common
descent, though newcomers might be adopted into them. No man who had
sufficient kin could want for protection and status.

That the herdsmen were nomads does not mean, of course, that they wan-
dered indifferently according to their fancies. Each larger grouping — nor-
mally, what we call a tribe or some division thereof — possessed its own re-
cognized pasturing grounds; even within these grounds, major movements
might be made in massed armed groups. But the camel nomadism of Arabia
was less closely tied to fixed grazing grounds and seasonal itineraries than most
nomadism elsewhere. There remained a good deal of leeway for unpredict-
able wandering by smaller groups and even individuals, which helped give a
tone to the Bedouin social ways. Each tribe, almost each clan or group of
families, was sovereign; led by its chief, chosen partly for his family descent
and partly for his personal wisdom, each group defended its own gra ling rights
in its own area, or attempted to better its position at others' expense. Each
group had to take decisions in which all might participate and which
repeatedly could mean life or death for all adult males of the group.

Such a society rejected authoritarian political forms and based itself in-
stead on individual prowess and prestige and on close lineage group loyalties.
Besides leadership in fighting, the chief might serve to arbitrate disputes or he
might be custodian of the group's sacred symbols; but others might serve in
either of these roles. In any case, he had no authority to coerce the acceptance
of his position by any family. Every man was free in the last resort to depart
at will with his dependents. In the absence of any common court of justice,
intergroup restraint was maintained by the principle of the retaliatory blood-
feud: an injury by an outsider to any member of a group was regarded as com-
mitted against the whole group by the whole group to which the outsider be-
longed; the injured group's honour required that it exact from the other group
in retaliation an equivalent — normally an eye for an eye, a life for a life
(though commutation in goods might be accepted) — or more, if the injured
group regarded itself as above the level of the other. But if the retaliation were
regarded by the other group as excessive, it in turn was honour-bound to re-
taliate again — till the feud could somehow be stopped. 3

3 W. Robertson Smith, Kinship and Marriage in Early Arabia (London, 1885) chaps.
I and II, conveniently presents the nature of the Arabian tribes and their genealogies.
Modern Bedouin life is studied from the viewpoint of its political potentialities — in any
period — by Robert Montagne, La civilisation du desert (Paris, 1947I.

150 THE ISLAMIC INFUSION

From time to time, especially in association with some sedentary power,
such large agglomerations of tribes could be formed acknowledging common
leadership, that the chieftain at the top, normally basing himself at least in
part on urban revenues, could take on something of the role of a king. But such
kingship was founded, even so, on tribal presuppositions. Unlike a tribal
chieftain, the king could have orders carried out by his own agents with re-
latively little regard to group sentiment; but his power was based on the
lineage ties of a tribe, which ultimately he could not flout. Such kingdoms
were precarious in their power at best.

The camel nomads were the elite of the more arid parts of Arabia. In addi-
tion to their camels, they often had horses which they pampered and used for
specialized raiding purposes; more humble but more economically useful were
sheep and goats. But the dignity of a tribe was likely to be in inverse propor-
tion to its dependence on the smaller animals; for sheep-herders had to stay
near the agricultural lands, and necessarily found themselves at the mercy of
those more mobile than themselves. The pure camel-herders, more mobile and
resourceful than either agriculturists of the oases or other pastoralists, felt
that they had, and they were widely conceded to have, greater prestige than
any other people of the area. Even when more settled groups did not actually
descend from Bedouin who had taken over the oases by superior force, they
looked to Bedouin traditions as the most honourable to follow.

The camel nomads called themselves 'Arab. And the earliest appearances
of the term seem to connect it with camel-herders as such. But, presumably in
consideration of the nomads' prestige, the term 'Arab came to mean not only
Bedouin proper, still herdsmen, but also settled Bedouin who would still have
camels but who lived from the date palms and grain of the oases or were en-
gaged chiefly in commerce. Hence the dominant population of the peninsula
came to be called 'Arab and their language (a form of Semitic differing some-
what from the Aramaic of the Fertile Crescent) 'Arabic'. (Subsequently, 'arab
has been used in Arabic itself in several senses, some of them quite extended.
Here we will use 'Arab' to refer to any person whose parental tongue is de-
rived from the Arabic of the Peninsula. 'Arabic' is used with reference to the
language itself, and 'Arabian', of course, only with reference to the Peninsula
proper, not to the Arabs generally.) 4

The economic life of the Bedouin tribes, while immediately a matter of
herding or (when they settled in oases) of agriculture, was ultimately depen-
dent on an extensive system of trading and raiding. The herding groups
always depended on agricultural groups for essential food or equipment,
notably for grain and dates to supplement camel's milk and meat ; beyond this, a
more long-distance trade brought in luxuries like wine or skilled singing (slave)
girls, in which all tribes desired to share. Thus the nomads became involved in

4 For the several ways in which the term 'Arab' has been used among scholars —
which must be kept distinct when one reads their works — see section on usage in
Islamics studies in the Introduction.

MUHAMMAD S CHALLENGE 151

the commerce between the Mediterranean lands and the Southern Seas. Tribes
near trade routes furnished camels for transport, escorted traders, or even
traded on their own. Others shared in the booty by raiding the more fortunate.
The occasional towns, accordingly, as nodes of the trade, formed a focus of
tribal aspirations. They enjoyed an influence based on wealth and prestige,
which was not, however, necessarily expressed in any political domination.

The Arabs in international politics

The Arabs were proud of their independence, blazoned in glorious tribal genea-
logies. But once the Bedouin society was fully developed, then both politically
and economically their life was constantly entangled with that of the great
empires around them, on whose commerce their own trading and raiding de-
pended. As the map of Arabia shows, Bedouin Arabia lay between three agri-
cultural lands: the Iraq, Syria, and the Yemen. ('Uman, relatively infertile,
cut off from the main mass of the Arabs by the wastes of the Empty Quarter,
and giving only on the south Iranian deserts, counted for little.) Syria and the
Iraq formed the main portions of the Fertile Crescent, the long-standing home
of the Semitic cultural traditions. The Yemen, since about iooo b c, had been
the site of agrarian kingdoms of a Semitic language (south Arabian) and of a
culture related to that of the Fertile Crescent but distinctive. (The kingdoms
there had left a more recent memory of greatness than the Semitic kingdoms
of the Fertile Crescent, a memory cherished by tribes with Yemeni associa-
tions.) The Yemen throve partly on agriculture, which had been declining,
and partly on trade, which had been growing more important over the mil-
lennia, between the Southern Seas and the north. Like the Fertile Crescent, it
was Christian and Jewish in religion, but it had a more important pagan sector
than had survived in the Fertile Crescent.

Each of the three lands was connected with what may be called a political
hinterland — a highland region which tended, in the sixth century, to dominate
it. Behind the Iraq lay the Iranian highlands, homeland of the Sa.sa.nian em-
pire. Syria had long been ruled by an empire based in the Anatolian high-
lands — and more generally, in the Greek-using peninsulas of the north Medi-
terranean. For the Yemen, the Abyssinian highlands were less important. The
Abyssinian citied culture itself had derived originally fairly directly from that
of the Yemen, though it had struck independent roots and developed a dis-
tinct language and its own dominant Christian church. But the Abyssinian
monarchy — commonly in alliance with the Roman empire — had cultivated
commercial and political pretensions which had culminated in an occupation
of the Yemen by Abyssinian forces, which had ruled there autonomously until
overthrown by the Sasanians at the end of the century.

The Bedouin were playing an ever larger part in the life of all three sur-
rounding lands, and hence of the empires which dominated them. The greater
part of the Yemen, by the sixth century, seems to have been Arabic-speaking:

152

THE ISLAMIC INFUSION

the south Arabian language died out soon after the rise of Islam, though
traces of the dialects have persisted. There were many Arabic-speakers also in
Syria, and they played a role in the Iraq. This was partly due to the normal
tendency for population distant from cities, and less subject to the great
scourges of life on the agrarianate level, to fill in the recurrent gaps in the more

Towns and tribes in Arabia in the time of Muhammad

settled population. It was also due to the active initiative of the Bedouin
Arabs as traders and as soldiers. The trade routes north and south along the
west Arabian littoral were of long standing; in late centuries they were no
longer controlled by Yemenis or by Aramaic-users based in Syria, but by
Arabs of Bedouin background. Newer trade routes across Arabia in the middle

MUHAMMAD S CHALLENGE 153

and around the northern fringes — skirting conventional customs stations and
perhaps river pirates — were also in Arab hands, and important enough to
enter Roman-Sasanian treaties. When Muhammad was growing up, much of
the transit trade between the Indian Ocean and the Mediterranean basin was
passing through Arabian overland routes. This may have been due partly to
the long sequence of wars between the Roman and Sasanian empires, which
encouraged the enterprise of neutral merchants who could bypass troubled
frontiers; it was made possible, however, by the full development of the tech-
nique of camel transport, which began to replace the donkey or the ox-drawn
wagon for long-distance hauls. Moreover, Arabs were becoming mercenaries
and allies in the great power struggles.

All three lands had their political projections into the Bedouin Arab com-
plex. At the end of the fifth century, the Yemeni kingdom was supporting a
great tribal grouping in central Arabia, under the lead of the Kindah tribe;
presumably it should have acted for Yemeni interests in trade and perhaps
politically to balance Roman and Sasanian power, but just then the Yemen
was occupied by the Abyssinians. The Kindah power broke up almost im-
mediately. But soon the Sasanian and Roman empires were each sponsoring,
and subsidizing, their own Bedouin tribal kingdoms. On the southern borders
of Syria, the Romans gave high title to the Ghassanids, partly expecting pro-
tection from raiding by other tribes, partly as an Arab arm in their power
struggle with the Sasanians, in which the lands south of the Fertile Crescent,
like the mountainous region north of it, offered opportunities for gaining com-
petitive advantages. The Sasanians financed the Lakhmid kings at Hirah near
the Euphrates, at the end of an Arabian trade route to the Iraq; these con-
trolled tribes far to the south and the northwest and had great prestige even
in central Arabia. The Lakhmid was the strongest and most enduring of the
Bedouin kingdoms, till after 602 the Sasanians took direct control of Hirah
and its military resources. But even apart from these kingdoms, Arabs were
serving as mercenaries at least in the Roman armies. The Lakhmid forces seem
to have been sufficiently well placed to have helped decide a contested suc-
cession to the Sasanian throne. The development of camel nomadism had
reached a point where it was impinging importantly on the surrounding lands.

The Arabs had surely been in touch with each other, over the whole of
Bedouin Arabia, from the time when nomadism was fully developed. But it
was under the stimulus of the international competition that was pouring
money into the Arab kingdoms, and doubtless partly because of the prosperity
of the Bedouin trade routes, that in the fifth and sixth centuries the rudiments
of an Arabic high culture grew up. The first Arab merchant-kings had used
Aramaic as their formal tongue. Ghassanids and Lakhmids were proud to use
Arabic. There was surely always tribal poetry, but in these centuries the
poetry reached a peak of formal specialization.

The supreme cultural expression of the tribal life — again whether among the
nomadic Bedouin tribes or in the towns — was a highly cultivated body of

154 THE ISLAMIC INFUSION

poetry in a standard all-Arabian form of Arabic, sometimes called mudari or
'classical' Arabic. Very early, poetry was connected with the transient Kindah
kingdom. Later the most important poetic centre was the Lakhmid court,
which rewarded panegyrics grandly. The patterns of metre and of sense in
these poems were highly stylized and the individual poets were given great
recognition as tribal spokesmen. Despite a tendency to kingly patronage of
distant poets, even the Lakhmid chiefs shared in this tribal mood. Each new
poem was soon carried throughout Arabia by professional reciters, especially
if it bore some relation, as it commonly did, to the greater intertribal feuds.
For at this time — perhaps not unrelated to the wider commerce — a network of
feuds and political struggles tended to involve the whole of Arabia in a single
political complex, if a rather incoherent one. 5 The Arabs had their common
sagas and common heroes, and their common standards of behaviour worthy
of a Bedouin.

The Meccan system

The Quraysh Arab tribe at Mecca made a special place for themselves in this
Arab society. The most important trading centre of western and central
Arabia was Mecca in the Hijaz. It was at the junction of two major routes.
One went south and north, through the mountainous Hijaz from the Yemen
and the Indian Ocean lands to Syria and the Mediterranean lands; the other,
of less importance, went east and west from the Iraq, Iran, and the central
Eurasian lands to Abyssinia and eastern Africa. As compared to Ta'if and
other central Hijaz localities in the same area, Mecca was relatively un-
promising; unlike Ta'if, it had no great oasis — that is, sufficient underground
water was not tapped there to form a watered agricultural area. It had suffi-
cient water to satisfy many camels, however; it was protected by hills from
Red Sea pirates; and it possessed a respected shrine to which pilgrimage was
made.

Some generations before Muhammad, under the leadership of one Qusayy

5 On the condition of pre-Islamic Arabia, and especially of the Hijaz, see Henri
Lammens, Le Berceau de V Islam: V Arable occidentule a la veille de Vhegire, vol. I, Le Ctimat
— Les Bedouins (Rome, 1914), especially part III, 'Les Bedouins'; and Henri Lammens,
'La Mecque a la veille de l'hegire', in Melanges de V Universiti St. -Joseph,^ (Beirut, 1924),
97-439.) In using Lammens, the reader must beware of Lammens' over-scepticism;
often Lammens' doubts leave the evidence hanging in mid-air, and sometimes he
exaggerates (e.g., in introducing modern commercial terminology). The standard
historical work will now be Jawad 'All, Ta'rfkh al-'Arab qabl al-Isldm (6 vols., Baghdad,
1951-57). Frants Buhl, Muhammeds Liv (Copenhagen, 1903), translated as Das Leben
Mitkammeds (Leipzig, 1930), has a lucid and judicious chapter describing relevant
conditions, 'Forholdene i Arabien ved Tiden for Muhammeds Optraeden'. Sidney
Smith, 'Events in Arabia in the Sixth Century', BSOAS, 16 (1954), 425-68, is useful
on dating. See also Giorgio Levi della Vida, 'Pre-Islamic Arabia' in The Arab Heritage
ed. Nabih A. Faris (Princeton University Press, 1944) — a skilful summary of what we
do and do not know about the history of pre-Islamic Arabia as a whole; note that the
translator has misrendered some words so badly that they can be misleading to the
unwary reader at important points.

MUHAMMAD S CHALLENGE 155

and with the aid of tribes along the route to Syria, a tribe had been brought
together, called Quraysh, to take over the springs at Mecca and the shrine
from other Bedouin. The Quraysh were organized on Bedouin principles, with-
out a king or any other municipal institutions beyond the clan councils; they
used an assembly of notables of all the clans for non-binding consultation.
The threat of blood-feud guarded the peace. But ever since Qusayy's time, the
Quraysh had maintained solidarity (not without some clans gaining a position
of more influence than others) and had made effective use of their resources.
They controlled the north-south trade and grew rich by it. To do so, they had
also to win a secure diplomatic (and warlike) position among the tribes of the
Hijaz, which was then bolstered with a financial position — leading tribesmen
became their creditors. For they engaged not only in the long-distance trade,
but in local trade in western Arabia; they had fostered the pilgrimages (and ac-
companying fairs) made at certain seasons to Mecca itself and to a neutral
spot not far away ('Arafat), as well as other markets held in the region. They
became the dominant partners in an alliance with the Thaqif tribe of nearby
Ta'if, where leading Meccans had summer houses. In the course of all this,
they had acquired prestige as a tribe of dependable and independent honour. 6

Their position was institutionalized in religious forms. The fairs took the
form of pilgrimages, and to protect the traffic at those times the Meccans
established sacred truce months, four a year, which a large number of tribes
observed. To settle the times of the sacred months, the Meccans maintained a
calendar of their own, equally widely used. Internally, their solidarity was
maintained through the worship at the Ka'bah, a rectangular building which
formed the object of the Meccan pilgrimage (hajj) . This worship seems to have
embodied a somewhat unusual development of the Arab paganism.

Among a world of minor spirit beings (jinn), the Arabs distinguished a
number of more serious divinities, often as protectors of particular tribes;
each was associated with a shrine at some given locality, a tree or a grove or a
strangely formed rock (or sometimes with a sacred stone or other object carried
ceremoniously by the worshipping tribe). The greatest divinities were likely
to be associated also with stars. People supplicated or propitiated them with
special rites in view of some worldly hope or fear. Back of these active divini-
ties was a vaguer figure, Allah, 'the god' par excellence, regarded as a Creator-
god and perhaps as guarantor of rights and agreements which crossed tribal
lines. But, as with many 'high gods', he had no special cult. 7

' Julius Wellhausen, Reste arabischen Heidentums (2nd ed. Berlin, 1897; also in
Skizzen und Vorarbeiten, Heft 3), section 'Mekka, der Hagg, und die Messen', discusses
the reasons for Mecca's importance. For the commercial position of the Quraysh and their
political relations, see Lammens 'La Mecque a la veille de l'liegire', especially chaps. I-
III and XIII-XV. (But beware exaggerations.) Irfan Kawar, 'The Arabs and the
Peace Tready of ad 561', Arabica, 3 (1956), 181-213, brings out some suggestive points
about the Arab commercial context of Meccan activity.

7 On the pagan Arab cults, see Wellhausen, Reste arabischen Heidentums, especially
the final section; but note that the conception of a supreme creator-god need not be
explained linguistically, for such figures are widespread.

156 THE ISLAMIC INFUSION

In the Ka'bah were gathered the sacred tokens of all the clans of Mecca; it
thus merged their several cults into one. Qur'anic testimony shows us that the
Ka'bah was presided over by Allah, presumably in his capacity as guarantor
of agreements among the tribes, and hence as guarantor of the pilgrimage as
well as of the agreements among the Meccan clans. If Allah still had no special
cult, at any rate in Mecca he thus came into special prominence. (It seems that
even Christian Arabs made pilgrimage to the Ka'bah, honouring Allah there
as God the Creator.) The special role of the Ka'bah as shrine of joint pagan
worship was not limited to the Meccans. In addition to the sacred tokens of
the Meccan clans, other tribes in alliance with the Quraysh were encouraged
to bring their tokens and fetishes there, so as to join in a common sacredness.
A number of the more active divinities seem to have received special honour
at Mecca: notably three goddesses (Allat, al-'Uzza, and Manat) who were
widely worshipped among the Arabs and in particular had shrines in the neigh-
bouring districts with which the Meccans had close relations. Worshippers
honoured the Ka'bah by circling it a fixed number of times on foot, and
touching the sacred stones built into it: particularly the Black Stone in one
corner. Near it sprang a sacred well, called Zamzam. It was the centre of a
hallowed area, extending all round Mecca, in which fighting was taboo even
when it was not a truce month.

Mecca was located approximately equidistant from the three spheres of
power around Bedouin Arabia. Midway between Syria and the Yemen, it was
about equally distant from the long arm of Sasanian power to the northeast.
Perhaps only at such a distance from the agrarian lands could so thoroughly
independent and Bedouin a system have arisen. A major task of the Quraysh
— on their own behalf and perhaps also on behalf of the tribes of their allies —
was to maintain the independence of their zone. Both Romans and Abys-
sinians had made expeditions to the area (the Sasanians had not, and it is clear
that the Quraysh rather favoured the Sasanians over either Rome or Abys-
sinia). When a Jewish dynasty came to power in the Yemen, Abyssinia, as
Christian ally of Rome, had finally intervened to overthrow it, on the pretext
of halting persecutions of Christians, but perhaps also on account of the
Sasanian and anti-Roman sympathies shared by many Jewish groups, even in
Syria. The Abyssinians seem to have sent expeditions as far north as Medina
against Jewish settlements along the trade route. The Abyssinians in turn
were ejected by Sasanian forces, evidently gladly received in the Yemen. In
Muhammad's lifetime, an attempt by one of the Quraysh to forge links with
Byzantium (and possibly rise to power in Mecca himself) was frustrated by
Meccan insistence on neutrality.

The corollary of maintaining political neutrality was to maintain neutrality
among the religious allegiances that disputed among themselves the lands
from Nile to Oxus and the Fertile Crescent in particular. This was not neces-
sarily easy. Bedouin Arabia was a prime mission field offering opportunities to
casual merchants or to solitary monks. Arabs were keenly aware of those

MUHAMMAD S CHALLENGE 157

venerable agrarianate high cultures in which they as Bedouin had little share,
and in particular of the confessional communities that played so prominent a
part in urban civilization. Some Arab tribes had even adopted for themselves,
to some degree or other, one of these religious allegiances in place of the not
very lively tribal paganism of their ancestors. We may surmise that the rest
of the Arabs could not long resist conversion to one or another such religious
allegiance. Perhaps only the want of a single allegiance that would automati-
cally command adherence had allowed them to linger.

In the area around Bedouin Arabia the confessional traditions, all of the
Irano-Semitic monotheistic type, proliferated as diversely as anywhere in the
Nile-to-Oxus region. Most widespread was Christianity, which in a variety of
mutually hostile forms prevailed in the Mesopotamian plain (Nestorian and
Jacobite Christianity), in Syria (Jacobite, Armenian, and Chalcedonian Chris-
tianity — the latter being the official Christianity of the Roman empire, later
split into Roman Catholic and Greek Orthodox), in Egypt (Coptic and Chalce-
donian), and in Abyssinia (Coptic). Judaism and Christianity were also es-
pecially strong in the Yemen in the far south. On the east coasts of Arabia,
Zoroastrianism was also important. In most of Bedouin Arabia, and especially
in the Hijaz, the mountainous area of the west in which are Mecca and Medina,
none of the confessional allegiances had yet become prevalent. Bedouin
Arabia, never incorporated till then into the great agrarian empires that had
risen and fallen north of it for so long, was a still pocket of paganism, where
the commonest form of religion was the old worship of local and tribal spirits.
But all the main religious allegiances were represented there; even in the
Hijaz there were some Christians and a great many Jews. When Muhammad
preached a religion of one God, of prophets, and of Hell and Paradise, the
terms he used could be understood by many Arabs, even among the pagans.

Yet Muhammad may have been in the one place where paganism was still
most vital. As the camel nomads began to play a role in the agrarian lands
and in international politics, the Quraysh of Mecca were playing a role not
only influential but politically and religiously unique among them. In contrast
to the precarious pyramiding of tribal agglomerations with a king-like chief-
tain at the top, they had been able to base a reasonably effective political
order on the solidarity of one tribe, and its prestige. And this was cemented in
an equally independent religious system, likewise based on Bedouin ways, and
equally neutral to all the confessional religious allegiances. The Meccans seem
to have offered the only effective Bedouin-based alternative to assimilation to
the settled cultures. 8

8 Joseph Chelhod, Introduction a la sociologie del 'Islam: del ' animisme kl'universalisme
(Paris, 1958) is not about the sociology of Islam as a religious tradition, but about the
development of religious consciousness in Muhammad and his compatriots, in their
social context. Chelhod builds upon Lammens' work very suggestively, stressing the
evolution of Mecca itself in the Hijaz; unfortunately, his racialism leads him to mis-
conceive the course of subsequent Islam; and even on the proper subject of his book,
his arguments are mostly very tenuous. See the excellent study by Gustave von

158 THE ISLAMIC INFUSION

Muhammad becomes a prophet

Abu-1-Qasim Muhammad b. 'Abd-Allah 9 was a substantial and respected mer-
chant in Mecca. He had grown up an orphan, under the care of an uncle in
unprosperous circumstances; but he was of an established family (the Banu
Hashim), members of the Quraysh, the ruling tribe of Mecca. He had shown
his competence as a trader in the service of a well-to-do widow, Khadijah,
some years his senior, whom he subsequently married. By her he had four
daughters (and evidently sons who died in infancy), whom he was able to
marry into prominent families. 10 He was known in Mecca as al-Amin, 'the
trustworthy'.

In his thirties, if not earlier, Muhammad seems to have become preoccupied
with questions of how to live a serious life in truth and purity. He apparently
listened to all who had something to say about the meaning of human life in
this world and he meditated intensely in periods of retirement in a cave (on
Mount Hira.') outside the town. He did not dissociate himself from the rites
and customs of the Quraysh, which indeed continued dear to him. But he
sought something which they lacked.

We shall be talking a good deal about religion in this work, for it pervaded
every aspect of Islamicate culture. I must make clear my point of view on it. ' '
It may be said that the religious impulse is ninety per cent wishful thinking.
Wishful thinking is, indeed, rooted deeply in us. Unlike other animals, human
beings live by their illusions: our very words, it has been said, point to what is
in fact not there. Human beings alone are artists. Over and beyond the im-
mediate stimulus and response, we want every moment to make sense in some
larger whole which our lives form: people cannot stand living with sheer absur-
dity. If we refuse to make a conscious choice of what sense to make of life, we
are told, we will in practice adopt some pattern of sense unconsciously and
without consideration. Hence even intelligent people may persuade themselves
to believe almost anything that seems to make hopeful sense of life. And since
life is largely a tissue of miseries, we are under pressure to discover some sense
which will give the misery a positive meaning. The logic of wishful thinking,

Grunebaum, 'The Nature of Arab Unity before Islam', Arabica, 10 (1963), 5-23, on the
pre-Islamic Arabs and the conditions that allowed Mecca to play a special role among
them.

9 The abbreviation 'b.' stands for ibn, 'son of; for a fuller explanation, see the
section on personal names in the Introduction.

10 On Muhammad's connections and the status of families in Mecca, see W. Montgomery
Watt's Muhammad at Mecca (Oxford, 1953), chaps. 1. 2 and II. 3.

" I am personally a convinced Christian, of the Quaker persuasion, but neither here
nor earlier do my general formulations on the nature of religion represent Christianity as
such. If they represent anything, it is the sort of considerations that have been developed
in the modern discipline of religion studies in the works of such scholars as Rudolf Otto
and especially Mircea Eliade; not without influence from the anthropological tradition
(e.g., Bronislaw Malinowski, Paul Radin), the sociological (e.g., Emile Durckheim), cer-
tain psychologists (e.g., Carl Jung), and philosophers (e.g., Ernst Cassirer, Wm. James,
Albert Camus),

MUHAMMAD S CHALLENGE 159

then, is not to be despised: that if something is possible (though proof or dis-
proof is unattainable), and if it is desirable, then it may be presumed true
until disproven.

But what is remarkable about human beings, in distinction from other
animals, is what we have done with our illusions — with our free imaginations.
Artists deal, in a sense, with illusion; but if they are disciplined, they can
evoke reality by way of the illusion. In religion also a disciplined imaginative
response can touch reality. The component of sheer wishful thinking in religion
is large, but is still not the whole of it. What is most interesting about religion
is not the wishful thinking as such, but the creative insights that come along
with it, which open new possibilities of human meaningfulness and expres-
sion. To identify these expressions of religion in particular, the observer must
sometimes penetrate into motives and implications not immediately apparent
to outsiders in the words used in religious discourse. But this need not mean
distortion. What serious and intelligent persons over many generations, and
in preference to many available alternatives, have held to be significant rarely
turns out, on close investigation, to be trivial.

Despite Qusayy's measures, the prevailing religious climate in Mecca was
still not far removed from the Bedouin paganism round about. Relations with
fetishes or with deities were chiefly on the basis of bargaining — for this offering
I give you, lord, you will give me that favour in return. This was little removed
from magic. Lots were cast at the Ka'bah to foretell fortunes, and vows and
sacrifices were made to assure successes. A sense of loyalty certainly there
was; but there seems to have been little higher moral challenge. Even the
special presence of the Creator-god Allah did not carry far. No meaning
or goal was added to a man's life other than what he already had as a
tribesman.

But a different sort of spiritual attitude was also represented at Mecca, an
attitude associated with monotheistic religion. Foreigners who were settled in
Mecca, or just passing through, were attached to a number of Christian and
perhaps Jewish communities and to other monotheistic communities of a
'Gnostic' type such as were then common in Syria; Meccans were aware that
such religion was supported by the great empires, and knew that some im-
portant Arab tribes adhered to it en bloc. Apparently none of the representa-
tives of monotheism whom Muhammad knew in Mecca was very well-versed
in his faith. The terminology used in the Qur'an, as well as other details, re-
flects traditions from a variety of religious communities but presupposes no
intimate understanding of any of them. 12 Nevertheless, even an ill-trained ad-
herent of any monotheistic group could convey at least the possibility that a

12 Whatever one thinks of the provenance of the Qur'an, its language is designed ex-
plicitly to be intelligible to Arabs. Hence we may gather from it what sort of concepts
could be expected to be intelligible to Muhammad and his contemporaries. That is, we
can learn from it with assurance what ideas were circulating in Mecca at the time. It is
only those ideas of which the Qur'an would make use.

l6o THE ISLAMIC INFUSION

man's life must be measured by larger than tribal standards — that his actions
counted as those of a human being, not just a tribesman; that to be good they
must accord with the nature of the world as a whole, not just with personal or
tribal interests; and that accordingly only at a summing up of the world's
whole history could any man's life be seen in its proper light.

There were a few other men of the Quraysh in Muhammad's time who were
attracted to monotheism; they seem to have worked out, perhaps each for
himself, some sort of private faith — later tradition called them the 'Hanifs'.
From these, Muhammad was distinguished by a crucial event. Toward the age
of forty, during one of his retirements in Mount Hira.', he heard a voice and
saw a vision which summoned him to offer worship to the God who had created
the world, the God of the monotheists, to Allah whom the ordinary Arab
honoured but had no cult for. Encouraged by his wife, Khadijah, he accepted
the summons as coming from God Himself. Thereupon, he received further
messages which he interpreted as divine revelation, and the worshipful recita-
tion of which formed a major element of the new cult. The messages collec-
tively were called the Qur'an, which means 'recitation'. For a time, only his
wife and a few close friends shared the cult with him. But after some years the
messages demanded that he summon his fellow Quraysh to the worship of God,
warning them of calamities to come if they refused. From a private monotheist
he was to become a prophet to his people.

We know far less about Muhammad than was once supposed. On the face of
it, the documentation transmitted among Muslims about his life is rich and
detailed; but we have learned to mistrust most of it; indeed, the most re-
spected early Muslim scholars themselves pointed out its untrustworthiness.
But we do know a great deal more about him than, for instance, about Jesus.
The evidence about Jesus is almost exclusively contained in the four Gospels
and in a letter by Paul. The more they are analyzed, the less dependable the
Gospels prove to be. Even the recorded sayings of Jesus can be shown to have
been heavily edited at least in some Gospels. As to the personal spirituality of
Jesus we have only the thinnest evidence. We may surmise that he was sincere;
but we are already in the realm of conjecture when we try to say what he was
sincere in. We tend to choose as characterizing him those episodes that are
most touching, or most distinctive and unlikely to have been common stock.
This is probably a sound instinct, but it is dangerous from a scholarly view-
point; at the least, it must be checked on the basis of other texts from the
period, which are only just becoming available. In the case of Muhammad,
though we must use a large amount of conjecture, we can base it on reasonably
objective scholarly principles. We can rely on the text of the Qur'an itself as
direct evidence — 'though that text is habitually ambiguous in any concrete
references it makes. To interpret the Qur'an, we are forced to resort to reports
collected several generations later; but even among these, we are not entirely
at a loss: we can probably rely on those reports which can be shown not to
grind the axe of any particular later party, provided such reports fit reason-

muhammad's challenge 161

ably well into a coherent picture that emerges from them all as a body.
And most important, we can often rely on the background detail which the
reports take for granted as known to all. Hence, though what I have
to say about Muhammad is largely conjecture, yet it can be responsibly
offered.

About Muhammad's call, we may say this. First, Muhammad accepted the
summons to the new cult: that is, he himself believed. This in itself was a
decisive act of faith. And then he did more: he accepted the role of prophet to
his people. This acceptance required not only unwavering faith in the validity
of his cause, but high courage; for it necessarily brought on him the scorn and
ridicule and mistrust of most of the men about him. For them, such a claim
was at best absurd — and at worst likely to be a cloak for private ambition,
perhaps connected with some Roman plot to control the Hijaz trade through a
local puppet, such as the Quraysh had had to resist before. Muhammad's
positive response to what he found himself confronted with, then, was his
great creative act. Before he summoned others, he himself had accepted the
consequences of his faith and staked his life on it.

A prophet is one who speaks for a god — who utters whatever messages from
the god are divinely laid upon him (not necessarily, of course, nor even pri-
marily, messages regarding the future). The impulsion to speak as a prophet
has been variously felt; it has ranged from institutionalized rituals, in which
abnormal physiological states are induced out of which dark words come, to
the expression of consciously personal insights by 'inspired' poets. Muham-
mad's standard for prophecy was, in principle, the experience and action of
the old Hebrew prophets. But he knew nothing of them directly. His own
experience was evidently very personal.

He found himself gripped in a distinctive physical condition and therein
becoming conscious of ideas which he did not recognize to be his own. The
physical stress seems to have been sometimes sufficiently violent that he
required wrapping up and he then sweated profusely; sometimes it was far
milder. (On the basis of some aspects of the detailed Muslim descriptions, the
moments of revelation have been very conjecturally likened to epileptic
seizures.) At such times he might be unconscious or at least abstracted. The
form in which the ideas came seems to have varied still more than his physical
condition. On certain occasions, presumably at the very start, we have the
evidence of the Qur'an that there was not only audition but ocular vision;
notably, according to reports, a waking vision of a gigantic being on the
horizon — on every horizon to which Muhammad turned his eyes — who spoke
to him the words he must say. Usually, however, there were simply auditions :
Muhammad heard words spoken or — in milder cases — some sort of tinkling
sound, to which, on his arousing, a meaning was attached.

The words which he then uttered were written down with care — he used a
number of 'secretaries' — and were memorized by the pious. They were re-
tained especially by the 'Qur'an-reciters', followers of his who specialized in

l62 THE ISLAMIC INFUSION

reproducing and teaching the whole of the constantly growing Qur'an. 13 The
various bits of revelation were arranged in surahs {'chapters'), some of which
represented a single revelation but most of which were added to from time
to time — often new passages being inserted in the midst of old.

A variety of incidents show that neither the occasion nor the content of the
revelations was under Muhammad's conscious control. After the earliest rev-
elations, there was a long period when he received none at all and became
severely depressed, doubting the truth of his own call — he was supported,
above all, by Khadijah's continuing reassurance. Later, Muhammad came to
expect a revelation at need. Accordingly, he was badly embarrassed on at
least one occasion when for days a required divine decision failed to come — an
experience he interpreted as designed to humble him.

The experience of being gripped as by an outside being, as well as the
ecstatic rhyming prose in which especially the earlier revelations took form,
seemed to the Arabs like the similar outbursts of a soothsayer, kdhin. These
men were regarded as possessed by jinn, sprites, who put their utterances into
their mouths. On a higher level, the respected poets, though they suffered no
state of seizure and what they uttered was in a far more developed literary
form, were regarded also as inspired by jinn. The words of both soothsayer
and poet were regarded as preternatural and possessing hidden powers.
Muhammad was at great pains to distinguish himself from either soothsayer
or poet; to insist it was no transient and irresponsible jinn that possessed him,
but a cosmic representative of the Creator-god Himself, an angel. (Eventually
he identified the angel as Gabriel.) Muslims have made a point ever since of
denying the Qur'an to be poetry in the technical sense, shi'r, as practiced by
the ancient Arab poets.

Monotheism and personal moral responsibility

At first it does not seem to have been clear that the new cult was incompatible
with the existing cults in which the Quraysh took part. In the new cult, por-
tions of the Qur'an were recited periodically to the accompaniment of bowings
and prostrations in honour of Allah. This was called the salat; as a form of
adoration it was reminiscent of Syrian Christian practice. Just as even Chris-
tian Arabs could take part in the hajj pilgrimage, the first Muslims, adding
their special practices, need not otherwise have made any notable break with
Quraysh customs. But from the first, the new cult set off its devotees as

15 The Qur'an-reciters (Qari') are sometimes called 'Qur'an readers' in English, but this
is misleading. They were often illiterate and in any case were not engaged in 'reading' the
Qur'an in the modern sense of taking in the meaning of a written text before one. (Unfor-
tunately, there are still Muslims who speak of 'reading' the Qur'an — or other text! —
when they mean merely declaiming it: uttering the appropriate sounds, with or without
sight of the printed page, without necessarily attaching any meaning at all to the
individual sounds uttered. In this sense, some claim to 'read' Arabic without knowing a
word of the language.)

muhammad's challenge 163

pledged to a new vision of life. The early portions of the Qur'an contain
numerous moral injunctions, urging purity, chastity, and generosity. The
specific moral ideals were in no case unprecedented and rarely departed from
moral norms upheld, in principle, in the older Bedouin society. (The Qur'an
made no attempt to lay down a comprehensive moral system; the very word
for moral behaviour, al-ma'ruf, means 'the known'.) What was new was the
conception of the place of these norms in a man's life.

In the Qur'an, the immensity of the human situation is brought out in the
descriptions of the Last Day, thus (lxxxi, 1-14):

When the sun shall be darkened, when the stars shall be thrown down, when
the mountains shall be set moving, when the pregnant camels shall be unten-
ded, when the savage beasts shall be stampeded, 14 when the seas shall be set
boiling, when the souls shall be coupled, when the buried infant shall be asked
for what sin she was killed, when the scrolls shall be unrolled, when the skies
shall be stripped off, when Hell shall be set blazing, when Paradise shall be
brought near, — a soul shall know what it has produced.

There is a moral condemnation of infanticide here, but not by way of intro-
ducing a new commandment nor even of reinforcing an old one. Rather, fe-
male infanticide, which was a natural consequence of the tribal emphasis on
males and of its disregard of the individual as such, is pointed to as showing
what sort of thing the godless soul 'has produced' — the quality of a life which
is without God.

In the Qur'an it was early made clear that human beings face a fundamental
moral choice. They cannot hover half way. On the one hand, they may choose
to stand in awe of their Creator and accept His moral demands. In this case,
God, in His mercy, will guide those who are faithful, making them upright
and pure. Or human beings may, on the contrary, turn away from their
Creator ; becoming absorbed in their private wishes of the moment, and praying
the various godlings for success in them. In this case, God will likewise turn
away from them, and they will become wicked, petty men and women. For a
human being cannot choose to be pure at will (a sad fact too readily experi-
enced!); he does not control his own ways, but can achieve moral purity only
by the power of God. 15 The fundamental choice, then, appears in the Qur'an
as overwhelmingly crucial: to turn to God and worship Him, or to turn from
Him to one's own desires. All else in the moral life will follow from this choice.

" Stampeded by fear, as Richard Bell suggests — literally, 'driven together'. The
scrolls, further on, are of course the records of persons' lives. I have based my renderings
on A. J. Arberry's but have departed from him freely so as to stay closer to the direct
simplicity of the original. (For Bell, Arberry, and other Qur'an translators, see the
section on translations of the Qur'an in the Selective Bibliography.)

" This account of predestination in the Qur'an seems to answer the tenor of the whole,
Maurice Gaudefroy-Demombynes, Mahomet (Paris, 1957), P- 357< suggests that Qur'an
xcn, 5-13, may express this idea succinctly. Certainly the usual interpretation of it
makes it improbably thin in moral content.

164 THE ISLAMIC INFUSION

Muhammad was convinced that this choice determined the whole worth of
a person's life. As did the other monotheists, he believed that this fact would
be made inescapably manifest in a final cosmic catastrophe, when the world
would be destroyed and all human beings would be visibly judged by God
Himself, those who had already died being restored to life for judgment. Then
those who had been faithful to God would be rewarded with all good things
the human heart delights in while those who had turned away would be
punished with all evil things the heart dreads. The Qur'an painted both the
rewards and the punishments in vivid colours — the blessed would dwell in
beautiful gardens with delicious fruits and charming damsels; the damned
would burn frightfully in fire, swallowing nauseous refuse. The likelihood of
the great final catastrophe was supported by descriptions of lesser catastrophes
that had overtaken individual peoples which had rejected the summons of
prophets sent to them; for instance, the people of Noah. Doubters were re-
minded that such a catastrophe, and the resurrection of the dead for judgment,
were entirely within the power of the God who had made the world and who
had formed each man and woman in the womb in the first place. But the
coming of the great Day was assured for Muhammad, finally, by its having
been announced in the messages sent to Muhammad and to all previous
monotheistic prophets as well. The same revelation which insisted on the
choice between the commands of the Creator and one's own desires also
warned of the final Judgment in which the choice would be vindicated;
Muhammad could not doubt any portion of the total message. 16

Accordingly, Muhammad insisted on the moral responsibility of human
beings. Life was no matter of play, it called for sober alertness; men dare not
relax, secure in their wealth and their good family and their numerous sons —
all these things would avail nothing at the Judgment, when a person's own
personal worth would be weighed; humans must live in constant fear and awe
of God, before whom they were accountable for every least deed. To be sure,
God is merciful: if a person were truly turned toward Him, slips could be for-
given in view of his human weakness; but a carelessness which neglected God
Himself for the transient delights He had given would not be forgiven.

The Qur'an puts the human situation in powerful images, drawn from
Biblical and Talmudic lore but reworked to express the vision of Islam. God

16 Very often, scholarly interpreters of the Qur'an have stressed the source of various
notions which appear in it. For our purposes, it is more useful to look for their meaning
within the Qur'an. The Qur'an could speak only in terms of the language and the concepts
which Muhammad and his followers already possessed with which to receive its message.
These were limited, on the whole, to what was to be found in the monotheistic traditions.
To understand a given reference, one must take note of the assumed context first: e.g., to
see the meaning of references to the Last Judgment, we must recognize that the notion of
the Judgment as an event in time via,s given by the monotheistic tradition. Then one must
see what the Qur'an does with this — what direction the notion is pushed in. It is only this
which can yield its meaning, morally and humanly. The same, of course, holds for all the
Qur'anic tales, whether Biblical or not. Their deviation from the Biblical form is relevant
primarily as pointing up the message they are to carry.

MUHAMMAD'S CHALLENGE 165

offered to the heavens the trust of keeping faith, and they refused; He offered
it to the mountains, and they said they were not strong enough; but human
beings undertook it. When the angels heard that God was creating mankind,
they objected. Why should God place in the earth creatures which would
simply fill it with injustice and bloodshed? But God insisted He knew what He
was doing. He taught Adam the names of all things (we would say Adam re-
ceived the faculty of rational discrimination), then He challenged the angels
to tell what things were; when they found they had to be taught this by Adam,
they acknowledged God was doing something beyond their understanding.
Indeed, God insisted that they all bow down and do obeisance to Adam and
all did so save Iblis (Satan), who was too proud. Thereafter, Iblis was permit-
ted to tempt humans to evil, and those who did not sincerely turn from him to
God would succumb. Then God drew forth all future generations of mankind
from the loins of Adam and confronted them with the demand, 'Am I not your
Lord?' They acknowledged Him each one and were tucked back in, to come
forth in due time to be tested, whether they would maintain the faith
or no. 17

The responsibility to obey God is thus imprinted in human nature, the
Qur'an tells us, but by people's carelessness the truth is forgotten — unless
warners come from God, prophets (nabi) who will bring to humans messages
from God such as Muhammad brought, whereby men and women may be re-
minded of their duty. The figures appealed to by the various monotheists had
been such prophets, bringing to their several peoples the same reminder of
their duty toward God. Many of them had performed wonders, deeds beyond
normal human powers, which authenticated their messages by showing that
they were supported by the same Being who performed the wonders of creation
and Who alone could be expected to produce wonders of a truly high order.
Thus Jesus, with the permission of God, gave life itself (to clay birds, as told in
an apocryphal gospel) , a wonder especially distinctive of the Creator and there-
fore unmistakably marking His intervention. Muhammad himself claimed
no wonder but the Qur'an itself; this, however, he regarded as undeni-
able. He challenged any man to produce its like; and (like any great creative
work) it has in fact proved inimitable. Nevertheless, it is clear from the
Qur'an that the divine message ought to be acknowledged without need for
any such evidentiary miracles. Those who are blinded by their delight in
transient things, and hence subject to the suggestions of Iblis (who can whis-
per his temptations to the heart, though he has no power of himself to mis-
lead us), will reject any prophet, however well evidenced. But those who have
guarded themselves from such blindness will recognize the truth as soon as
they are reminded of it by the warnings of a prophet. The prophet's mission,
therefore, is in the first instance simply to utter the warnings as God gives them
to him.

17 Qur'an, xxxiii, 72; ii, 30-34 (28-32); xxxviii, 71-85, xvii, 61-65 (63-67); vii, 172
(171)-

l66 THE ISLAMIC INFUSION

At least latent in Muhammad's message from the beginning was the idea
that there is only one true object of worship, only one God; that all other god-
lings are not just secondary but absolutely false, and their cults wicked. For a
time, Muhammad may have preached the new cult without insisting on the
overthrow of any of the old cults; once he even tried to find a place for a cult
of the greater, Meccan goddesses as intermediaries, subordinate to that of
Allah; yet, before long, insistence on the exclusive cult of God alone became
the central dogma of Islam. 18

As Muhammad confronted both his opponents and his own followers with
the new message, and as its implications were lived through, it became clear
that no concessions could be made. If Judgment was to be total and final, the
Judge must be utterly transcendent, incommensurate with all that was to
come under His judgment. There could be no intermediaries, no half-gods.
And if people genuinely made the fundamental moral choice of turning to God
for His guidance, they should not turn back to the petty cults which were there
to serve only their lusts. The various monotheists agreed in imposing such a
ban, and the ban became the crucial test of whether a person had become a
muslim, whether he was undertaking the obligations of faith. The essential
step in joining the Muslim community came to be abandonment of idolatry
(shirk, literally association of something else with God) ; that is, any cult of
beings other than the Creator-god.

The monotheistic communities whose representatives Muhammad looked
to as worshippers of the Creator-god had developed complex theories about
the divine realm. The elements of an elaborate supernatural apparatus seem
to have played a considerable part in Muhammad's thought also, particularly
at first. The Qur'an presupposes acquaintance with various cosmic figures and
objects. It mentions, for instance, angels, the Spirit (Ruh), the Word (Qawl),
the Command (Amr) ; several heavens, the divine Throne, and heavenly books
— the latter recording not only human destiny, but men's and women's deeds
in the world and the divine message itself (from which the Qur'an comprises
only excerpts). 19 On the other hand, Muhammad seems not to have been
acquainted with the more central doctrines of the New Testament — those of
divine incarnation, of suffering, sacrificial love, of redemption; nor to have
been acquainted with the major literary prophets of the Hebrew Bible, such
as Isaiah or Jeremiah. Nevertheless, he seems to have become less concerned,
in time, with the supernatural apparatus, and to have centred into an outlook
far more in harmony with the great Hebrew prophets than with the sec-
tarianism, often of Gnostic type, whose terminology and sacred tales he knew.
It is as if he had been led back of the popular imagery that was visible to him,
in the chance representatives of monotheism he encountered, to the grand
themes of the old prophets whose personalities had not seized the popular

18 On the 'Satanic verses' and on the early evolution of Muhammad's mission, cf. Watt,
Muhammad at Mecca, chap. V.r.

" On the Qur'anic cosmology cf . Gaudefroy-Demombynes, Mahomet, pp. 2Q2ff.

muhammad's challenge 167

imagination, but whose basic insight still formed the solid groundwork under
all the later luxuriant overgrowth.

Muhammad, with the Qur'an, presented a potent challenge to everyone at
Mecca: a challenge to rise to a level of personal moral purity such as it had
occurred to few to dream of. He presented it as a real possibility for human
beings, indeed a necessity if they were not to risk offending the very structure
of the cosmos in which they found themselves. And he presented it in a con-
crete, tangible form in which, by an act of will, they could adopt the new ideal
practically.

Muhammad founds a religious community

When Muhammad began to preach publicly his new cult, particularly when he
began to oppose the old cults, most men of the Quraysh naturally ridiculed and
opposed him; but he won many converts, especially among the younger men.
Some converts were slaves or tribeless persons, but most were from the less
powerful Quraysh clans, and a number were among the less well-placed youn-
ger men in the very top families. With the development of Mecca as a com-
mercial and financial centre, the moral standards of Bedouin society no longer
served well. Though the Quraysh seem to have kept the dangerous custom of
feuding under control at least within Mecca, a type of economic inequality
had arisen between man and man which threatened tribal solidarity and in
any case undermined the Bedouin ideal of generous manliness in which wealth
was a welcome but relatively transient distinction. In Mecca, as the indivi-
dual began to act more freely in his own private interest, the tribal expecta-
tions came to fit less well. Particularly those who were disadvantaged in the
new, more individualistic pattern welcomed a moral conception which could
restore something of the older moral security in a form adapted to indivi-
dualistic, commercial life. To this end, Islam was highly appropriate. Muham-
mad's creative act of accepting prophethood thus found a public capable of
responding to it.

It was not only his message that mattered, however; Muhammad's per-
sonality backed up the message. He was able to convert and hold the loyalty
of diverse sorts of outstanding and able men. As the chart of Muhammad's
relatives shows, he won their respect on the most intimate level. Two youths
of his own household, whose conversion may no doubt be ascribed to his pri-
vate influence, grew into exceptionally strong leaders. His uncle, Abu-Talib,
had entrusted one of his sons, 'Ali, to Muhammad to raise; 'Ali as a boy may
have been the first male to accept Muhammad (i.e., the first person after
Khadijah) ; as a man he proved a powerful warrior, winning in his own person
an almost fanatical loyalty from many men. Almost as early a convert was
Zayd b. Harithah, a freedman of Muhammad, whom he adopted for a time
as a son, and who was later a trusted general for Islam.

One of the earliest converts from other clans was Abu-Bakr b. Abi-Quhafah,

tfi

IS

ft

rt

,G

<L>

en

tf

1?

H

3

a

B

R

U

rrt

.£)■
I

T3

a

3
o

M-t

3

>>

s

ccj

Tt

3

rv

3

13

- a, .

p
■

<p

M

.3
-co "

<

■d —

£ B

£>

.3 2

O <nj

eg i^h

3

!a|

a -t!

jtj g

^^ rt

<cii

3

<<

<3 ■

M
<!

3
<ccj

<

X
<<

<

i

a
■ n

<

■a

3

«1

5

i

<

n

<
w

<P

M

m

3
A

•a
d

B

nj
H
3

3*

6

«

9'

<C3 •

* :

3 t

ca-

<cs a

.3 3
cS ,Q

n<5

■ «!

<<

IS
X
H
P

.3

nj

3

TS
S
<S

B

•3
.3
+j

-"3
«

fi
B

■§

3

- >-.

N

«
«!
>■
>-

it

«

i-l

»

H-J

a

*"■*

o

fi

+j

T3

d

ni

CL)

B

B

B

r3-

o

. CO

■d o

3

3 + J

§

.2 3

.g

tt rt

B -=>

CD 4->

4)
i — i

B »

O

CU

u

•8 >

+>

CO .^

CJ

A "*

cti

+-> CD

+J

^-1
O
ft

B

+" 13

* ,H

2 el

3

B ^

cS

O

to to

-d

1) CD

>1

O +3

_rt

B °

z

ft

<

P=

3 a"

til

p

O

CO

§*

w-

13

.-tt

to T3

S B

*+-!

ft

o

oS

CO

o

B B

cu

B

3

•r-t

to" ^>

K

nS

V

O 3

<
in

&

a s

muhammad's challenge 169

a well-to-do merchant (though not among the very wealthiest), who acted
somewhat as Muhammad's faithful lieutenant, devoted his wealth unhesit-
atingly to the cause, always showed a cool good judgment and an admirable
impartiality, and won abiding general respect among the Muslims. 'Uthman
b. Maz'un, also an early convert, contrasted with Abu-Bakr in having inde-
pendently become something of an ascetic monotheist before Muhammad's
preaching and seems to have continued, even while loyal to Muhammad, as
something of a leader on his own; he is listed as chief of those Muslims who
went to Abyssinia to escape persecution. 'Uthman b. 'Affan (who was to be-
come a caliph), was a retiring and deeply pious member of one of the best
families of Mecca. 'Abd-al-Rahman b. 'Awf was a clever businessman; it is
said that later, on arrival penniless in Medina, he refused an offer of property
from a Medinese friend, took a trifling item to the market place, and quickly
traded himself into a good supper — afterward becoming quite wealthy.

'Umar b. al-Khattab was impulsive, almost fiery, but intensely committed
to whatever won him. He is said to have been talking of killing Muhammad
out of zeal for the Meccan godlings one day, when he was told for the first
time that his sister had herself become a Muslim; going to her in a rage, he
found her engaged in reading a portion of the Qur'an — and on reading it him-
self he was impressed, and converted; thereafter he forced the Quraysh to let
him perform Islamic worship publicly at the Ka'bah itself, which was not
tolerated before, and all the Muslims could do so likewise. His uprightness was
never doubted; on occasion he chided Muhammad himself. His sternness is
said to have been such that levity which was tolerated in Muhammad's own
presence was suppressed if 'Umar appeared. After Muhammad's death, he
governed the Muslims with genius during their greatest conquests.

Islam soon became more than a private cult; it became an issue dividing
the town. Muhammad demanded that all Mecca join his cult and abandon all
others, on pain of divine chastisement. This demand created something of a
dilemma. Though in principle Muhammad was but one of the faithful, distin-
guished from the others only as recipient of the Revelation and as specially
charged to warn the rest, in practice he could not but be more. The lines of
commitment which tied together the new group, and which tied the new group
as a whole to Muhammad, had an emotional and moral strength which could
well rival that of a weakened tribal solidarity. Muhammad, as prophet, could
readily seem the authorized interpreter of the moral implications of the mes-
sage which he brought, and therefore have an authority over the faithful.
Muhammad and his followers always distinguished between the Revelation
from God and Muhammad's own decisions. For years some of his strongest
and closest followers were inclined to come to their own conclusions as to what
the new faith implied in given situations. Nevertheless, Muhammad showed
himself a man of good personal judgment and when the demands of faith be-
gan to conflict with accepted standards at Mecca, he naturally came into a
position of leadership among the faithful.

170 THE ISLAMIC INFUSION

After some years of the preaching in Mecca, the Qur'an revelations began to
recount the experiences of former (usually Biblical) prophets with the peoples
to whom they were sent. As in earlier monotheism, the divine challenge was
taking historical form. The cosmos of the Qur'an was intensely human and
even social; if a single sequence of historical unfolding does not emerge as in
the Hebrew Bible, despite its adumbration in the first part of surah 11 (for the
Qur'an rejected the idea of a chosen people), yet we feel a strong sense of com-
mon human destiny. In these historical stories, the prophet came to figure as
something like the head of a community of faithful, set off against their
opponents; the story of Moses was a favourite. The stories, of course, reflected
Muhammad's own circumstances in Mecca and therefore gave divine sanction
to what was becoming increasingly evident: that for the Quraysh to accept
the new cult of Allah would mean accepting not merely the moral demands of
the Creator, but also the political lead of Muhammad himself. In the nature of
prophecy, a transcendent universality appealing to the individual conscience
was inescapably linked to concrete relationships with a particular human
group and its leader.

The tribal forms of Meccan life allowed no more for municipal than for
monarchical institutions. Hence when Muhammad preached repentance and
the worship of the God of the great empires round about, there was no govern-
ment to jail him as a traitor or even as a public nuisance; nor for his followers
to seize and reform. Instead, the struggle between the reformers and their
opponents took the form of personal manoeuvres and of family and clan op-
positions. In such a struggle, the Muslims, distributed as a minority among
many clans, were at a disadvantage. There was a certain measure of personal
persecution. The weakest of Muhammad's followers, especially slaves, who had
no clan to retaliate against any who might do them harm, were molested by
the more zealous enemies of Islam, and occasionally tormented painfully;
Muhammad himself was sometimes insulted in ways hard for an Arab to bear.
One of Muhammad's uncles, called in the Qur'an Abu-Lahab (the 'flame man',
as doomed to Hell) was one of the most vehement in fighting the movement.
But the Banu Hashim as a whole, led by Muhammad's uncle Abu-Talib,
loyally stood by their clansman Muhammad, even though most of them did
not accept his cult; threat of retaliation from them, accordingly, prevented
any direct personal injury to Muhammad. The same seems to have held good
for most of the better-placed faithful, though some may have suffered consider-
able loss through financial pressures, for which retaliation was not provided
under the Bedouin code of honour.

Two years after the mission was made public, some Muslims (about eighty)
began to emigrate to Christian Abyssinia, across the Red Sea, where they
could expect (and did find) asylum among monotheists. A few became Chris-
tian and stayed there, but most subsequently returned to Mecca or (later) to
Medina when Muhammad was established in that oasis. Their motive was in
part to escape from persecution but they may also have been expected to form

MUHAMMAD S CHALLENGE 171

a base for some sort of wider plan envisaged by Muhammad. 20 Such a project
would be the first instance of Muhammad's attempting measures on a political
plane toward a solution of the dilemmas presented by his becoming a prophet;
in any case, the adventure illustrates how quickly his followers were becoming
an autonomous group, with its own destinies as such.

Perhaps three years after the mission was made public, the other clans of
the Quraysh, failing to persuade the Banu Hashim to cease protectingMuham-
mad, joined in a boycott of the Banu Hashim, refusing to have commercial
relations with them. The Banu Hashim held out against this for two or three
years till there arose sufficient disunity among the Quraysh for the boycott to
be given up. But shortly after this, in about 619, Abu-Talib, Muhammad's
protector, died and Muhammad's position with the Banu Hashim became more
questionable. At the same time, he lost his wife Khadijah, who had been a
major spiritual support to him. Muhammad began taking active steps to find
a more satisfactory base for his work.

A visit to neighbouring Ta'if proved fruitless; he was abused and driven
away. Returning to Mecca, he was able only with difficulty to assure himself
protection from families outside the Banu Hashim (now, it seems, led by his
enemy Abu-Lahab) so as to remain provisionally in Mecca. He seems to have
been greatly discouraged. He was gratified to feel that a number of jinn
(sprites) who happened to be listening as he recited the Qur'an were converted,
and he seems to have made a good deal of a dream vision which he was granted,
in which he visited Heaven, or possibly Jerusalem, city of the prophets. (This
vision was later greatly elaborated among Muslims, as the mi'rdj, and given a
central place in Muhammad's legend.) Muhammad's insistence on such things
seems to have scandalized some of his followers and to have contributed to
their defection during these years.

He may have regarded his mission, till then, as directed chiefly to his own
people, the Quraysh, just as other prophets had been sent to their own people
{though at all times any human being who was present was certainly included,
for the message was in its nature universal). Clearly, however, most of the
Arabs, being pagans and not monotheists, were equally in need of the message,
and now he preached to all who would listen at the great fairs in the vicinity of
Mecca. In 620, about a year after Khadijah's death, he met (at 'Aqabah, on
the pilgrimage) with a handful of converts from Yathrib (afterwards called
Medina), an agricultural oasis about two hundred miles north of Mecca; the
next year they returned with a few more and not only declared themselves
Muslims but pledged themselves to obey Muhammad in any wholesome com-
mand; in 622 six times as many were at hand to reinforce the pledge, pro-
mising Muhammad and his Meccan followers protection if they came to Me-
dina. In the same year, Muhammad sent most of his followers, rather more

20 For possible implications of the migration to Abyssinia, cf . Watt, Muhammad at
Mecca, chap. V.2.

172 THE ISLAMIC INFUSION

than seventy, to move one by one to Medina, and finally went himself. (This
move, or 'migration', was called in Arabic the Hijrah.)

Pressure against the Muslims had been mounting in the last year or so.
There was some sort of effort to prevent Muhammad's leaving at the last
when it was realized what was afoot — the Muslims subsequently, at least, inter-
preted it as a plot to kill him. Muhammad and Abu-Bakr, his lieutenant, fled
by night and hid in a cave till the search had slackened (away from their
protectors at Mecca, but not yet at Medina, they would presumably have been
fair game). Then they made the long trip secretly, till at the outskirts of
Medina the local converts came to meet them and receive Muhammad as their
chief.

The Ummah achieves autonomy

To the Medinese, such a step was a solution to pressing problems. Medina had
evidently been developed or restored as an agricultural oasis (notably raising
date palms) by Jewish Arab tribes; apart from their religion, these tribes
shared the same culture as the other Arabs, though presumably at least the
nucleus had immigrated from farther north, originally being Aramaic-speakers
from Syria. With the adoption of Judaism had come sufficient of the Jewish
law to provide good order. In recent generations, however, other clans had
settled there, still pagan, who had not adopted Judaism. Their Bedouin ethi-
cal system of 'honour' had plunged the settled clans into ever more intense
feuds; in Muhammad's time they had lined up in two main tribes, 'Aws and
Khazraj, who had come to a deadlock such that no man was safe outside the
limits of his own fields. Living among Jewish tribes, even the pagans at
Medina had come to recognize and no doubt to respect monotheistic religion.
In Muhammad's message, many saw an opportunity to adopt a monotheistic
cult of their own which would at the same time introduce a new type of moral
sanction and a neutral leader to represent and apply it as arbiter of the dead-
locked quarrels. Thus they added to the public which was proving capable of
responding to the Revelation. Even those pagans who were little interested in
the new faith could welcome a neutral arbiter to restore peace. 21

To Muhammad, the move to Medina was not merely an escape from an
untenable immediate position in Mecca. It was an opportunity to build a new
order of social life such as the development of his faith had more and more
obviously demanded. The cult of Allah as Creator demanded, in the first in-
stance, a personal devotion to moral purity; but personal purity implied a just
social behaviour: generosity to the weak and curbing the licence of the strong.

2 ' On conditions in Medina and the relations among Jews and pagans, cf . Wellhausen,
'Medina vor dem Islam', in Skizzen und Vorarbeilen, vol. 4 (Berlin, 1889); but his
assurance that most of the Jews must have come from Palestine seems based partly on
racialist presuppositions. Clearly, conversion to Judaism meant a sharper break with
Bedouin tradition than did conversion to Christianity, but descendants of converts were
capable of acquiring both settled habits and Talmudic learning.

MUHAMMAD S CHALLENGE 173

Moreover, it was fully recognized that a person's moral life is usually less a
function of his good resolutions than of the level of actual expectations around
him. It must be society and not just individuals that should be reformed. The
Qur'an makes it sufficiently clear that the new way is for everyone, not just
for moral heroes, by praising almost as much those who urge others to a virtue
as those who practice it themselves. The new life must be lived by a society at
large.

This called for more than exhortation to be good; it became clear that Mus-
lims could not rest content to be irrelevant deviants in a society founded on
contrary principles. Sooner or later, the challenge of the Qur'an was bound to
require the creation of a just polity as the natural outgrowth and context of
the personal purity it required. Obviously, no pagan who did not accept the
challenge could have remained neutral to such claims in the long run. What-
ever the personal circumstances, the pagan Quraysh could not have tolerated
a movement which attacked the principles of their social order and suggested
an alternative sort of moral sanction for behaviour and especially for social
authority. So long as the Muslims stayed as a minority in Mecca there would
have been deadlock at best.

Muhammad and his followers had been gradually emerging in Mecca as
something like a new tribal formation (possibly, on this level, analogous to
what happened when the Quraysh had been first gathered by Qusayy). Each
man retained his loyalties to his inherited clan; but (and here Qusayy's ex-
ample was departed from) he was coming to have overriding loyalties to the
new grouping, based not on family ties but on individual acceptance of the
faith which Muhammad preached. The word Muhammad used for the new
grouping was Ummah — a word which he had used of a people to which a pro-
phet had been sent (such as the Meccans) , but which now was applied to such
of that people as did respond to the prophet and so formed a new community
with him. In the negotiations with the Medinese Muslims, Muhammad claimed
explicitly an authority over the religious community which had become in-
increasingly implicit already even among his Meccan followers. He did not yet
make the same demands among non-Muslims, of course, but the political
autonomy of the Muslims allowed them to establish at least a certain level of
social expectations among themselves. Moreover, as hakim, judge-arbiter,
among even those Medinese who were not Muslims, Muhammad was able to
extend something of the new spirit outside the Ummah proper even before
the whole of Medina merged into the Islamic group.

The life of the new Ummah was to be marked by a pervasive new moral
tone, derived from the individual's relationship to God, but maintained (as
moral standards must be, save perhaps in the case of 'moral athletes') by the
expectations prevalent in the group as a whole and given form in their cor-
porate life. The new tone was contrasted to a moral orientation associated
with both Bedouin pastoral life and with the settled Bedouin pagans. These
had stressed, above all, individual and group pride and point of honour —

174 THE ISLAMIC INFUSION

pride in birth, pride in one's wealth or prowess, pride which led, when crossed,
to an unremitting, pitiless vengefulness; to a passionate and heedless (if some-
times magnificent) pursuit of self-centred, inherently trivial ends. Even those
who might prefer a different way of responding to life were dragged into the
pattern by the voice of public opinion, urging vengeance as the most practical
means of justice, and praising transient delights of drink and sex in which to
forget the pettiness and pointlessness of a life which time would sweep away.

To the heedless pettiness of men, which Islam presented as sheer ingratitude
(kufr) to their Creator, Islam contrasted humility, generosity, and a serious
purpose of carrying out the demands of God in a pure life. To the passionate
Vengefulness of men, Islam contrasted patient restraint and mercifulness.
Some of these virtues had, indeed, been recognized and extolled by the wiser
among the Bedouin, though humility and restraint in carnal pleasures had
not; but the accent in any case had been different. Islam was to reinforce in its
community life the godly virtues, provide more just alternatives to the relent-
less feud, and remove all persuasion to what were newly felt as vices.

The contrast was summed up in key words. The old heedlessness of God
was termed kufr, 'ingratitude* or 'denial', to be replaced by submission to Him,
islam; the old harsh passions were summed up as jdhiliyyah, to be replaced by
trust and faithfulness to God, iman. (In later times, kufr became a theological
term for whatever was incompatible with Islam, and Jdhiliyyah became a his-
torical term for the age before Islam appeared.) 22 And the new moral order,
like the old, was not to be merely a personal ideal but the effective norm of a
total, responsible, social environment.

When it was disentangled from the clans at Mecca, and so could form a
political unit of its own, the Muslim Ummah at first still took an essentially
tribal form. In Medina, Muhammad was the acknowledged commander of the
Muslims, both those of Mecca (called Muhajirun, 'emigrants') and those of
Medina (called Ansdr, 'helpers'). He was also, more generally, arbiter among
all the social groups at Medina. This position was established in a document
(sometimes rather grandiosely called the 'constitution of Medina') in which
the mutual obligations of the adhering clans were set forth, and all Medina
was included by way of clan alliances. But at first his primary role lay among
the Muslims as such and especially his own Meccans. His Meccans, the
Muhajirun, lacked resources when they arrived; they became the guests of the
Medinese Ansar, with certain of whom, to begin with, he paired them as
brothers. Almost immediately, he began sending the Muhajirun out to raid
the trading caravans of the Quraysh.

Raiding was, of course, the normally received procedure whereby the less
well-placed Arab tribes recouped their disadvantages from the more fortun-

" On the relation between the moral tone of the pagan Arabs and Islam, cf. Ignaz
Goldziher, 'Muruwwa und Din', in his Mukatnmedaniscke Siudien, vol. i (Halle, 1889);
Watt, Muhammad at Mecca, chap. Ill, 3, 4; Toshihiko Izutsu, The Structure of Ethical
Terms in the Koran (Keio University, 1959).

MUHAMMAD S CHALLENGE 175

ately placed. In leaving Mecca, the Muhajiriin had broken their ties with their
own clans there, abandoning claims to protection by them; they now formed,
in effect, a clan or tribe of their own; that they should raid (if they could)
those with whom they no longer had any agreements was taken for granted.
Muhammad, however, felt the step nevertheless required justification in terms
of the new moral orientation. He pointed out therefore that the Quraysh were
not merely declining to become Muslims themselves; they were actively op-
posing the divine order with their persecution of Muslims and their inter-
ference with the public practice of Islam. This was not just a private grief of
the Muhajiriin, but a public cause on which the eternal fate of others hung; it
was right to fight the Quraysh till they should no longer place obstacles in the
way of those who might otherwise become Muslims.

In particular, there were two outstanding motives for launching the raids,
which set them off from the normal Arab raiding. In the first place, they were
a means, important if not absolutely necessary, for Muhammad's own men to
gain an independent economic position at Medina, without which the life and
social order of the new community there must remain artificial. (Perhaps also
such raids, once they became effective, might form an outlet for the passions
of the Medinese, debarred henceforth from the old feuding; but it is doubtful
if Muhammad would have acknowledged such a motive as being independently
valid.) Secondly, the raids were to humble the pride of the Quraysh, perhaps
in anticipation of coming acts of divine displeasure — or even as part of those
acts, as it might appear when fighting came to be commanded in the Qur'an
itself. It may have been that Muhammad already had the aim of ruining their
trade and reducing them to a recognition that in Islam they were meeting
something bigger than they supposed, with which they must come to terms
even to survive in the present life.

The first raids were unsuccessful. The first success was that of a small party
which attacked (at Nakhlah, near Mecca) a caravan during a sacred truce
month, killing one man and bringing home the booty. It is unclear how far
Muhammad was directly responsible for the violation of the truce month, but
he may have anticipated it; the event proved to be a scandal at Medina, which
was calmed only by a revelation in the Qur'an that while violation of the
truce was bad, persecution of the faith was worse and justified the violation.
Muhammad then allowed the booty to be accepted. This act deepened the
breach between the Muslims and the Quraysh to a breach with the whole of
pagan Arab culture, and particularly the Meccan system, of which the truce
months were a primary symbol. The primacy of Islam over all old customs
was asserted ; in effect, no bond or tie of pagan society need hold, in the Islamic
community, unless explicitly acknowledged anew within Islam. In the follow-
ing years this principle was implemented consistently. Whatever crimes a man
had been guilty of, even against Islam, when he adopted Islam the slate was
wiped- clean. And just as a Muslim could not be punished for what had gone
before, so he could not profit by attachments with the past: he retained his

176 THE ISLAMIC INFUSION

own property, indeed, but could inherit nothing from a pagan relative. Thus
the Ummah of Islam was proclaimed wholly independent.

Muhammad's acceptance of the success at Nakhlah despite the qualms of
his followers may be thought of as a mark of courageous consistency in which
he discerned, without flinching, the complete implications of his mission and
carried them through as occasion presented itself. Had he compromised here,
he might well have been reducing his Ummah to the status, in practice, of but
one more competing tribe within a common pagan moral framework. In the
clarity and single-mindedness of aim here displayed surely lay much of his
genius. At the same time, the inherent dilemma in Muhammad's mission
found here concrete expression. Through the prophet, transcendent truth was
brought into men's and women's lives by being embodied in the work and
fortunes of a given human community, which was limited by given circum-
stances. It could thus take practical effect. But the raid at Nakhlah was not
simply a break with a superstitious custom, hallowed by pagan cults but in-
consistent with a new wider truth. Since there is no indication that the Mus-
lims had previously declared or implied that they would not respect the truce
months, it was also an outright act of treachery which Muhammad accepted,
and perhaps had to accept, as occasion for consummating the moral indepen-
dence of Islam. 23

Muhammad establishes a new polity

The success at Nakhlah seems to have encouraged a large turnout at the next
raid, in which many Ansar Muslims from Medina also participated. The cara-
van from Syria which was its object got by safely, but the raiders, some 315
men, found themselves at the wells of Badr face to face with a relieving force
from Mecca at least twice their size. Good generalship on Muhammad's part,
and presumably good discipline among the Muslims as well as high enthusiasm,
won them a smashing victory. Several leading men of the Quraysh, opponents
of Islam, as well as many lesser Meccans were dead or prisoners. Muhammad
regarded the victory of so few, over a numerous foe of such high prestige, as
being the result of divine intervention. It seemed to seal the independence of
the Muslim community and its ability to survive, and indeed to fulfill some
foretaste of the doom with which the Quraysh were threatened. In later years,
to have been present at Badr, at the first triumph of Islam as an organized
body, was like a patent of nobility. Later converts might be attracted by suc-
cess, but the men of Badr had been converted when Islam was weak and they
had held firm during its leanest years.

Henceforth the other Arabs, especially the nomads, must regard the Mus-

23 On moral implications of the Nakhlah raid, cf. W. Montgomery Watt, Muhammad
at Medina (Oxford, 1956), chap. 1. 3; also X.2.But the sacredness of the truce months
was surely not merely a matter of superstitious fear of retribution.

MUHAMMAD S CHALLENGE 177

lims as challengers to and potential inheritors of the prestige and the political
role of the Quraysh. Muhammad found himself in a position to attack some
Bedouin tribes which proved unfriendly; thus he gained booty and also a
freer hand against the Quraysh. And within a year a whole Quraysh caravan
had been captured. From then on, a primary activity of the community was
raiding and warfare, as its influence (and, later, conversion to Islam itself)
was extended. The warfare culminated eventually, but did not end, with the
surrender of Mecca itself.

Directly after Badr, Muhammad expelled the Jewish clan of Banu Qay-
nuqa' from Medina. This was in part an admission of defeat and a counter-
measure thereto. Muhammad had won his converts among the gentile clans
and his first authority was in the gentile sector of Medina. But he had always
expected that monotheists, whether Christians or Jews, ought to welcome his
message and give him support in his work among the pagans. But just as
serious Christians could not accept a timeless monotheism stripped of the
Incarnation, so most Jews could not accept a universalism in which their
history as the chosen people lost its unique significance. Moreover, Muham-
mad's versions of Biblical, Talmudic, and apocryphal Christian stories were
too patently incoherent, and sometimes garbled, to win the respect of those
who already possessed the older sacred books. There was little to encourage
them to hail Muhammad as prophet even to their pagan neighbours.

When Muhammad found that the Jews of Medina denied his prophethood
and ridiculed his misapprehensions of Biblical stories, he was deeply dis-
appointed. More, he was threatened. As interpreters of monotheism, the Jews
had undoubted seniority over the Muslims and were already respected in
Medina. As long as they challenged his authority, a single bad turn of fortune
might make his position untenable. Together with numerous smaller groups
of Jews, there were four important clans which were mostly or entirely Jewish;
of these, two held some of the best date groves in the oasis and a third, the
Banu Qaynuqa.', comprised the craftsmen and retail tradesmen in the market
at the heart of Medina. These three latter clans, at least, were in a sufficiently
strong position to hold aloof from Muhammad's arrangements. Muhammad
took advantage of a fracas between some Muslims and the Banu Qaynuqa'
(in which it chanced that a Muslim had begun the violence). He besieged them
in their strongholds till they agreed to leave Medina with their property, but
10 abandon their arms. (Expulsion or migration of whole clans, either within
ai> oasis or to great distances, had been of quite frequent occurrence in Arabia
and at Medina in particular, as an outcome of tension among neighbours.)
Numbering perhaps two thousand adults, the Banu Qaynuqa' migrated to
Jewish settlements further north.

The expulsion of the Banu Qaynuqa' consolidated within Medina the pres-
tige Muhammad had gained at Badr vis-a-vis the Quraysh and the Arabs at
large. He did not leave it at that. From this point on, at least, he was building
no longer just a new tribe but a more developed polity, in which both Muslims

178 THE ISLAMIC INFUSION

and non-Muslims at Medina were to be subsumed on the basis of their common
social life. While adherence to his leadership, at least apart from the Muhaji-
run, remained as yet voluntary, and from time to time even some professing
Muslims refused to go along with him, his position as general arbiter had taken
on weight. He now uncontestedly spoke for Medina as a whole. His chief rival
among the Medinese, Ibn-Ubayy, who had tried to intercede for the Banu
Qaynuqa', had been rudely discomfited. Islam, under Muhammad's leader-
ship, formed henceforth the ruling community in Medina and dissenters found
themselves at best tolerated. We do not know when any given provision was
inserted into the agreement called the 'constitution of Medina', but it was
appropriate to the spirit of this period that it contained a ban on any Muslim's
helping an infidel against another Muslim.

Not everyone who now proclaimed himself a Muslim was wholehearted.
Ibn-Ubayy became the leader of a group whom Muhammad called the
'waverers' {mundfiqun, sometimes rendered 'hypocrites'), who gave Muham-
mad trouble for several years. But he never gave them a chance to turn
against him openly. It was soon expected that all those who had been pagans,
at least, would now be Muslims; and Jews were expected to recognize Muslim
primacy. Meanwhile, the market place had become vacant, ready to be oc-
cupied by Muhammad's Meccans, who could thus gain a surer economic posi-
tion.

The new position appeared most clearly in the cult. In the quarrel with the
Jews it became clear that Islam not only was distinct from paganism but,
even within monotheism, formed an independent religious system, parallel to
and distinct from Judaism and Christianity. Whereas up to this point Muham-
mad had expected his cult to conform by and large to that of the Jews (for
instance, in praying toward Jerusalem and in observing certain fasts), now
the Muslim cult was set off markedly. Learning that Abraham was considered
the common ancestor of Israelites and of Ishmaelite Arabs, he pointed out that
Abraham was faithful to God though he was neither Jewish nor Christian,
having lived before either Moses or Jesus. Muhammad's cult was to be like
that of Abraham.

This decision sprang from two principles both very marked in the Qur'an,
which now came into their own. First was the principle of worshipping God
alone. Pure religion need be bound by no communal limitations — as Abraham
himself was bound by none. Muhammad proclaimed himself the ummi prophet,
that is, the prophet of those who had no sacred book — who belonged to none
of the established religious communities. This referred in the first instance to
his being an Arab. But it carried implications. The Irano-Semitic dilemma
of conflicting religious communities, which could become specially clear to the
thinking person in that mission field which was Bedouin Arabia, was to receive
its solution through a community that rejected the exclusivities of the old
communities and went back to the very font of the monotheistic tradition. In
principle, one cult was as good as another; wherever one turned his face, the

MUHAMMAD S CHALLENGE I79

Qur'an pointed out, there was God. A cult was instituted only to meet human
requirements; what mattered, as the Qur'an tirelessly stressed, was a per-
son's acceptance of God Himself, not his adherence to anything lesser (e.g.
surah vi, 160 ff.). This casual attitude toward any particular formulation of
cult or law is illustrated in the Qur'anic assurance that not only earlier revela-
tions were accommodated to their people's needs, but even within the present
revelation, God may set aside one verse of the Qur'an, and its injunctions with
it, and then a better will be given in its place.

Yet once instituted, of course, a command could not be ignored by those
who adhered to the new community; and to worship God properly, one had to
be in a guided community. The independence of the Meccan religious system
and the central place in it of the Creator-god Allah now proved a point of de-
parture for a new interpretation of monotheism that might transcend the com-
munal divisions of the older monotheistic traditions in a concretely practical
way. Abraham and Ishmael were naturally presumed to be the founders of the
chief shrine of Ishmaelite Arabia, the Ka'bah at Mecca, which was therefore in
origin dedicated only to the true God; the tribal fetishes there were subsequent
contaminations. One of the sacred stones near the Ka'bah was — then or later
— specially dedicated to Abraham (the Maqdm Ibrahim). The bowings and
prostrations of the formal worship, the saldt, were commonly done at least
three times a day in unison, normally at the prayer ground (mosque) at the
Prophet's home. Now Muslims were told to perform this worship in the direc-
tion not of Jerusalem but of the Ka'bah. They were also to look forward to
performing the hajj , the annual pilgrimage in the Meccan area, as a ceremony
instituted in principle by Abraham. Elements of the cult were derived also,
however, from the experience of the new community itself. The month of
Ramadan, in which the Qur'an is said to have been first revealed and in which
the battle of Badr was certainly won, was likewise instituted as a time of fast-
ing superseding the time of the Jewish fasts. Islam thus became ritually inde-
pendent of previous monotheisms.

The essentials of the new society were the new relations in it between
human beings and God and between one human being and another. But the
society was held together by the Prophet; his position was indispensable and
unique. The chief of a tribe, in view of his authority as commander in war,
and also as responsible for various obligations in maintaining the tribe's posi-
tion and honour, conventionally received a fourth of the booty taken in a
tribal raid. Muhammad, in a somewhat analogous position, received a fifth of
any booty, which he was to use for community purposes such as the relief of
the poor or the conciliation of new converts, and over the dispensing of which
he had complete authority. Also like some of the greater tribal chiefs, who had
numerous wives, Muhammad was authorized to have numerous simultaneous
wives beyond the four to which his followers were limited. Muhammad seems
to have used his marriages to cement political relations, so that this privilege
was, like the fifth of the booty, essentially a political one. (Only one of his

l80 THE ISLAMIC INFUSION

wives was a virgin when he married her, 'A'ishah, and in no case are social and
political reasons for a marriage not traceable.)

In harmony with their importance, his wives were to receive special respect
from the public. They were to live secluded, receiving their visitors (who were
many, on account of their supposed influence with Muhammad) only from be-
hind a curtain, rather than face to face. After Muhammad's death they were
not to marry again but were to be honoured as mothers of all the faithful.
Despite their political significance and the social hedge with which he had to
surround them, his wives meant much to Muhammad personally as well. Their
various quarrels produced his gravest personal emotional crises; the trace of
more than one of his marital complications is left in the Qur'an. None of them
ever took the place of Khadijah; in any case, he tried to treat them all on an
equalbasis without favouritism. Buthis best-beloved was Abu-Bakr's daughter,
'A'ishah, whom he married at nine years old and who was always the
liveliest of them all. With her he seems always to have unbent. Indeed, despite
his special position, Muhammad seems to have lived a quite simple and modest
life without any luxury, by and large accessible and affable with the lowliest,
delighting in laughter and in children.

An attack on Muhammad was felt as an attack on Islam, which he repre-
sented. Muhammad had deeply resented a number of individuals who had
abused him publicly, especially in verse. (Poetry was held in great respect by
the Arabs, being felt not only as the primary means of building one's own and
destroying one's opponent's morale, but even as something of almost magical
powers.) By existing Arab custom, Muhammad owed no obligation to persons
with whom he had no treaty. Shortly after the victory at Badr, he encouraged
some of his followers to assassinate, among these detractors, a certain man
and a woman to whom the assassins were closely related (and on whose ac-
count, therefore, they were not subject to blood-feud).

Beginnings of a new society and culture

Supreme within its own territory and spiritually independent, Islam could
begin to develop its own social order in earnest. It was scarcely as yet an
independent culture in most respects, to be sure; but increasingly many as-
pects of culture among the Muslims were differentiated in the new social con-
text. This was sometimes a matter of detail. Muslims were forbidden pork
(here Bedouin and Jewish feeling seem to have converged) and gaming and
intoxicants or at least the imported wine (this was in part a measure of social
discipline). 24 Most noticeable was a new system of assuring the security of the

2t The Bedouin may have had no objection to wild boar meat, but domesticated pig
was never suited to pastoralism and probably always had peasant associations. The his-
tory of prejudice about animals like the pig and the dog will not be satisfactorily eluci-
dated till we are able to see how probably diverse motives converged and were reinforced
by continuing circumstances; circumstances which can include ethnic pride, but cannot
include any inherent racial sentiment, too often invoked by scholars at present.

Muhammad's challenge 181

weak against the strong. Feuding among Muslim clans was forbidden and
equal penalties strictly proportioned to the offences were substituted, to be
exacted under the eye of Muhammad as God's representative. At the same
time, the financially weak were also provided for by the collection of zakdt, a
tax on possessions. This 'alms' tax grew out of a practice of alms-giving for
self-purification, zakdt; it was organized in Medina as the financial basis of
group life (along with Muhammad's fifth of the booty) as well as to serve in-
dividual justice: its proceeds were normally used at Muhammad's discretion,
like the fifth of the booty, either for the common cause or for the needs of the
poor, the traveller, and others in like case. Both the legally fixed criminal
penalties and the centrally distributed alms helped give individuals a status
independent of clan associations, and so could foster individualistic culture
traits.

Perhaps at the heart of any social structure is its family law. Certainly in
the Medina community it was in this field that the most explicit innovations
were made; so far as the Qur'an contains legislation, it largely regards family
relationships. The regulations were made piecemeal during the rest of Muham-
mad's life, but here again the tendency was persistently toward asserting in-
dividual rights on the basis of equality before God. We are not perfectly
acquainted with marriage practices among the pre-Islamic pagan Arabs, but
it is clear they varied greatly. In some cases the man acquired the woman
very nearly as property and brought her to live with his own clan; in some
cases the man seems to have retained but a casual relation to his woman,
who remained completely dependent on her parental clan. 25 What dignity
either a man or a woman had in the family relationship depended on status at
birth, family circumstances, and wealth.

At the centre of Muhammad's family arrangements were the Qur'anic rules
on marriage, which universalized one existing type of Arab marriage, with
modifications. The nuclear family — man, wife, and children — was stressed as
a self-sufficient unit, with every marriage given equal status at law. This was
largely achieved through strengthening the position of the individual adult
male. The man retained wide authority over the wife to the exclusion of either
his family or hers. The children were to be the husband's, who was responsible
for maintenance of wife and children. Inheritance was to be primarily within
the immediate family, not diffused through the clan. The degrees of relation-
ship within which marriage could not take place were stressed and even multi-
plied — with the effect that it was less easy for a married couple to be absorbed
by multiple ties within a wider household; thus relationship by fosterage

23 Watt has some suggestive, but not altogether proven, theses on the meaning of
traces of matrilocal practices at Medina, Muhammad at Medina, chap. VIII. 2. For a
more general discussion of family law, see J. Wellhausen, 'Die Ehe bei den Arabern',
in Nachrichten von der kdniglichen Gesellschaft der Wissenschaften zu Gottingen, no. 11
(1893), pp. 431-81. Corrections by Gertrude H. Stern, Marriage in Early Islam (London,
J 939). are not always reliable; in particular, her sampling could provide little evidence
for absence of polygyny, which is everywhere rare save among the wealthy.

l82 THE ISLAMIC INFUSION

(many of the Quraysh had their children nursed by foster mothers, that is,
wet-nurses) was made equal to relationship by blood. At the same time, the
integrity of the natural family was protected against the introduction of Ac-
tive relationships which might be independent of whether children had been
reared together: adoptive relationship, which the Arabs frequently created
between adults, was allowed no status at law. If a man formed more than one
sexual partnership — as sometimes happened among the wealthiest Arabs, as
elsewhere — each partnership must be given equal status with the first, up to
the number of four; each marriage, that is, must have the same tight-knit
character. More casual unions (with free women) were strictly forbidden, save
possibly in special cases.

At the same time, wives and daughters were given a stronger position than
they had had in those Arab marriages on which the Muslim form of marriage
was modeled. In Bedouin society the man had often given the bride's family
a 'bride-wealth', mahr, when taking his bride. In Mecca this had often been
given to the bride herself and this became the Muslim law. A substantial mahr
helped assure the woman's position. Under Islam, part of the male prerogative
as provider for the family was the right of divorce (though women also might
— at least later — sometimes initiate divorce) ; but if a man divorced he could
not regain his mahr. Nor could he make use of his wife's property during the
marriage, but must maintain her from his own resources. Both wives and
daughters inherited from their men, though sons, having to maintain new
families of their own, were given twice as much as daughters. The insistence
on the personal dignity of every individual, male or female, was illustrated in
the prohibition of infanticide, which had borne especially on infant girls.

In one case Muhammad and the Qur'an permitted an inequality of status
in the family to continue, though mitigated. As everywhere till recently, men
and women could own other men and women as property, though even such
slaves were allowed certain rights. Muslims were not permitted to enslave
Muslims, but outsiders could be enslaved. A man was permitted, in particular,
to take his female slaves as concubines, despite Muhammad's general disap-
proval of other than strict marital unions. Slaves were for the most part war
captives, commonly children who had been sold far away from their tribes and
so had no family when they grew up except that of their owners. Since in
Arabia no one could well exist without family and clan, slaves could not
usually expect to separate from their owners altogether; what they could look
to was an improvement of their status within the owner's family. This was
encouraged; the freeing of slaves was suggested as a common penance for
breaches of duty; then the ex-slave would be freely attached to his former owner.
But Muhammad did not compromise the principle of the solid nuclear family
by encouraging slaves to be adopted into it and receive the full right of sons.

One further aspect of family law received special attention in the Qur'an,
personal etiquette. The privacy of the home was to be respected and a modest
decorum was to be observed by both men and women outside the home.

muhammad's challenge 183

Though the rules laid down in the Qur'an were not very precise, they served
to support respect for individuals in their independent private lives. After
Muhammad's time, however, these rules became the starting point of a social
code of very different import.

The Qur'an and the community experience

As the Muslim community developed, the character of the Qur'anic messages
altered. The earlier portions of the Qur'an commonly have an ecstatic char-
acter, suggesting with great beauty the solemn majesty of the divine and
pointing up the awesomeness of the Revelation itself. These lyric solemnities
gave way to exalted but often rather prosaic exhortation and commentary.
The Qur'an served at once as the inspiration of Muslim life and the commen-
tary on what was done under that inspiration; its message transcended any
particular circumstances yet at the same time served as a running guide to the
community experiences, often down to seemingly petty details. It was filled
with repeated exhortations to support the community efforts, notably the
military excursions, and with regulations of community procedure, especially
in regard to marriage. Even particular crises were sometimes resolved by de-
cisions on disputed points or justifications of lines of action. At one point,
Muhammad's best-beloved wife, 'A'ishah, was accused of infidelity to him by
a faction hostile to her, in circumstances where no judgment, pro or con,
could rest on other than an estimate of her character. After a time of agonized
doubt, the Qur'an pronounced in her favour. But at the same time it brought
a rule requiring four witnesses in such an accusation — and upbraiding those
who had spread the cruel accusation without proof. A factional episode was
written into the Qur'an: and with it, moral observations on the episode which
carried beyond it.

The Qur'an did not generally initiate social policies as such. Here it was left
to Muhammad personally to act. The 'constitution of Medina', which settled
the position of the several elements in Medina when Muhammad established
himself there, was the work of Muhammad, not of the Qur'an. Time and again,
crucial decisions were left to Muhammad in his own person. Even at the crisis
of Hudaybiyah when, as we will see, the Quraysh stopped a much heralded
pilgrim expedition to Mecca, and Muhammad's wisdom was doubted by his
closest followers, the Qur'an did not intervene to dictate a course of action. It
concerned itself especially with individuals and with their individual con-
sciences. For instance, it did not order the burdensome expedition to Mu'tah
(toward Syria), which most Muslims would have liked to resist. But it dealt
individually with the cases of three slackers who had failed to join the expedi-
tion despite being sincere Muslims. Nevertheless, it steadily supported Mu-
hammad's policies, solving problems that arose out of them — for instance, the
distribution of booty — and above all it urged the supreme importance of
loyalty to the common cause as this was determined by Muhammad.

184 THE ISLAMIC INFUSION

Throughout the Qur'an, the transcendent point of reference in all this
human confusion was kept vividly before the mind, and a tone of grandeur
was maintained. In the whole monotheistic tradition, on its more populistic
side, ethics tended to be thought of in terms of the market — thus the pro-
tection of orphans was, in the first instance, protection of their property;
hence the Qur'an freely uses market terminology — partly by way of familiar
analogy (the faithful strikes a good bargain with God), but partly by way of
introducing the transcendent inextricably into daily life. But the manner of
using mundane references ennobled them. Even when the question of
'A'ishah's adultery is discussed, for instance, the very wording of the Qur'an
— word order and proportion, overtones, sonority — combines to keep the
discussion on such a level that the dignity carries the sordidness with it,
rather than the reverse; a quality that hardly can come through in transla-
tion. The wholeness of its vitality gives the Qur'an a certain self-sufficiency.
As the embodiment of an independent Islam, the Qur'an needed no supple-
menting from the older revelations. It became a many-sided, vivid, and
intimate possession mirroring the spiritual hopes and needs of each of the
faithful and above all bringing to ever new focus their common destiny as it
unfolded.

Because of its intimate interaction with the day-to-day destinies of the com-
munity, the Qur'an cannot be read as a discursive book, for abstract informa-
tion or even, in the first instance, for inspiration. The sequence of its bits and
pieces is notoriously often lacking in clearly logical order or development.
Even the stories it recounts come not as consecutive narratives but rather in
the form of reminders of episodes which are often presumed to be known to the
audience — reminders which point up the implications of the episode for faith
with little concern otherwise for continuity — -as if he who did not know the
story should ask someone to tell it him before approaching the Qur'an's com-
mentary on it. Hence many non-Muslims have found it a jumbled and inco-
herent mass, ridden with repetitions, and have been at a loss to fathom why
Muslims regard it as supremely beautiful. It must not be read through but
rather be participated in: it must be recited, as an act of self-dedication and
of worship. The Qur'an presents at every point one great challenge: to accept
the undertaking of faith. To recite it truly is to be accepting and affirming that
undertaking. Then its beauty can be responded to line by line and one will de-
light in the juxtaposition, whatever the immediate subject, of all its main
themes within any given passage. The repetitious phrases remind one of the
total context in which a given message must be understood: in even a small
part of the Qur'an, the act of worship can be complete.

By and large, the Qur'an did not emphasize the mysterious or the excep-
tional after the earliest period. It never lost the sense of majesty, indeed.
Even the relatively late 'light verses' of the Surah (Chapter) of Light, revealed
at Medina, illustrate an intense aspect of Muhammad's piety: they liken God
to an ethereal, supernally pure Light in images which can suggest a true mysti-

muhammad's challenge 185

cal experience. Yet the dominant tone of Muhammad's piety was to suffuse
everyday life with a powerful sense of transcendently divine requirements. In
the same Surah of Light, side by side with the most lyrical descriptions of
divine luminosity and of the desperate state of those who have lost divine
guidance, come simple exhortations, 'Perform the salat, pay the zakat, and
obey the Messenger; perhaps so you will find mercy. Do not think that those
who are ungrateful [to God] can frustrate [Us] on the earth; their sheltering
place is the Fire, a bad destination'. And then immediately come details to
encourage propriety in the household — bringing a sober, sensible discipline to
a community of ordinary people: 'You who are faithful, have those who are in
your possession [slaves] and those who have not reached puberty ask you
leave at three times [before coming in] : before the dawn salat, and when you
take off your clothes at midday, and after the evening salat, three times of
privacy [lit., nakedness] for you. Neither you nor they are at fault, apart from
then, when going about among yourselves. So God makes clear to you [His]
signs. And God is knowing, wise.'

In the opening surah of the Qur'an, the Fatihah, we find a typically sober
expression of the community's reverent hope and fear of God:

In the name of God, the Merciful, the Compassionate: Praise belongs to God,
Lord of all being; the Merciful, the Compassionate; Master of Judgment Day.
Thee we serve, on Thee we call for help. Guide us in the straight path, the
path of those whom Thou art bounteous to, not those whom anger falls on,
nor those who go astray.

Muhammad has been called 'the prophet armed'. This scarcely distinguishes
him from a number of other prophets, beginning with Moses. It is more helpful
to say that he was the prophet of the Ummah, of the confessional community.
The religious community had moved increasingly toward becoming the frame-
work of all high culture between Nile and Oxus. Despite his rejection of the
ultimacy of any one community's law — or perhaps because of the creative
freedom which this insight allowed him in building a new and purer com-
munity form, it was Muhammad's achievement to fulfill this communal ten-
dency at least in a single instance. His community at Medina formed nearly
the total framework of culture and society there. But the regional tendency
could be thus fulfilled at all only because Muhammad's community was not
designed simply to redeem the elect from the world, leaving to the Devil those
who failed to respond to its vision. It was designed to transform the world it-
self through action in the world.

But such a vision led inevitably to the sword. When those whose interests
will suffer by reform also wield power, maintaining jointly sufficient force to
put down any individual objections, reform will require changing the basis of
power. In the twentieth century, Gandhi has brought to the fore methods of
creative non-violence for producing basic changes in social power. But short

l86 THE ISLAMIC INFUSION

of these methods, a serious intention of social reform has commonly implied at
least readiness on the part of the reformers to use physical compulsion to meet
and overcome the compulsion used by those already in power. That is, it has
implied readiness to wage war — and to commit all the violence and deceit this
necessarily entails.

It is not just a Christian squeamishness, I think, that points to Muhammad's
military measures as a central problem in his prophethood. Every virtue car-
ries with it its own characteristic defects, every perception of truth is accom-
panied by its own temptations to falsehood. In any tradition, greatness is in
part to be measured by success in overcoming the peculiar failings which
necessarily accompany the peculiar excellences of the tradition. Christianity
has its own pitfalls. A peculiar test of Islam lies in how Muslims can meet the
question of war. In the loyalty and risk of warfare, a man used to find the
supreme virtue of dedication to a goal beyond himself to the point of readiness
to give up his life. But warfare — apart from the acts of individual injustice it
necessarily involves (since individuals are treated as elements in a mass) — is
at the same time the supreme expression of that claim to exclusive validity
for one's own position, which must be fatal to the open search for truth. Such
a claim to exclusivity has been, indeed, a standing temptation of all the mono-
theistic communities. Muhammad's prophethood, in fulfilling the monotheis-
tic tendency toward a total religious community, at the same time left his
community confronted with that temptation to a spirit of exclusivity that
went with any vision of a total community and that received appropriate ex-
pression in warfare. The resulting problems came to form a persistent theme
of Muslim history.

The Early Muslim State,
62^-692

Muhammad had created a new local polity, founded on his prophetic vision.
But almost immediately that polity took on far-reaching international dimen-
sions. Very soon it was contesting power within Arabia not only with the
Quraysh but with both the Byzantine and the Sasanian empires. Having
through these contests, made itself a general Arab polity, in the succeeding
two generations it extended its sphere over the neighbouring lands in monu-
mental struggles between the Muslim Arabs and the two imperial powers.
These the Arabs replaced, forming an established empire and organizing in
Arab and Islamic terms the life of the whole region from Nile to Oxus as well
as much of the west Mediterranean basin. The campaigns which created the
Arab empire were epic achievements.

But decisive struggles were equally required among the Arabs themselves
at each step in the formation of the empire. Only so could what started as a
loose association in the Hijaz, built around a charismatic individual, be trans-
formed into a massive and permanent state administering a complex agrarian-
ate civilization. At every turn, crucial decisions were made determining the
character which the Islamic polity was to take for the future. Ultimately, this
character, in turn, determined what impact Islam was to have on the society
it had conquered. It was the internal development which ensured that the
conquests should have any permanent significance.

Muhammad builds an Arab commonwealth between the Byzantine and Sasanian
empires

While Muhammad was creating a new social order at Medina, he was also
actively extending his influence beyond his chosen oasis. Indeed, it was this
militancy that both made possible and, at least in some measure, formed the
character of the Medina society. As can be seen from the chronology of Mu-
hammad's lifetime, once he was established at Medina decisive events followed
one another very quickly. The relatively slow preparation at Mecca bore fruit
with great rapidity, first at Medina itself and then at Mecca and in all Bedouin
Arabia.

The Quraysh were duly alarmed at the position Muhammad had achieved
at Medina after Badr. It was clear that Muhammad might possibly ruin their

187

ni

I

fl-

1=1

w

•s

cfl

a
o

< <cfl

ho W

-3, en

=d m

n, CO to

a)
fl

a
d q

03

J*

13 a

£•

~ 5

fl T3

fl "2

P '3

<3

3 ■

o

d
'3

<aj

en
w

ni

io I*

3 03
fl .3.
« ft

£3

°3

ftM

■a

S

o

en

d
o

to

o

CO

1 — 1

T)

O
fl

%

O

fl

u

S*

en

•h Si

g*«ciJ
H en

8 3
a a

c°
S <u
° A
<& *
- en
en tuO
3 d

■tl S o

s •--£
cs <u 9

d " 2

a .a j.

fl

(8

A
o o

a «

m

2 fl

^
<S -°

(2

en

cfl O _.

8 8£

«

8 H k

| O 03

<N 43 "2

00 ,3 d

ft

3

fl

h fl

*0

_, -=1
i-i +->

s

g

u

03

tj
d

3

-il

"p K
£? *

■I*

.1

S

T

OS

00 w k

U9 U> P

P

a

■a

3

43

rt

d

d

o

CO

■a

&

o

(A

03

■d

N

3

>

a

P

O

&

O ro

O

fl

8 £

03 tH
P< S3

0fl

O HH

5 d fl

O Cti TO

S-

1-1 .

el ■

o

CO

.3. >,

CJ 03

O J4

ft 4«

d

nj ..
^ l-i

B Q

'^

O g
oi i— >

en 2

Cj 03

ftH

d ih

cs

sp.SS

T

° 5

P r-t 2

i-i «e

o,d

(0 -^

d 03

rt >

o

ft
ft
o

d
o

03

o

-a

fl

O

d

>^

ft

3
O

d

tfi

en

ft
°

3

_03

P.
O

a

S?" d

d°|

<cS

+3

«j<

en

d

rH

«n)

as

?

•« <§

cn

CO

CM

U U3

u»

C

rH

ft«d

rH

O

CO

3 cn

CO

U

O

o
1-1

a

■a
<u

oj
fl-

I

I
fl-

.3

pq

o
r»
us

.0

^1

03 >?

■a

.0

I

3 03

•d

03

d

'3

42

ft

4j d

&0

■si

.» fl

§•6
o s

1 d°

S-t ^^

o d

C3 J

^ft

fl ifl

^

3

d
o

'-a

03
00 o
W C4

+3 fl il

31 i

^ 33

" 3
<fl

en a

fl

«J

fl"

d

o

O

fl

en

CO

to

<cfl

1

%

K

1

03

s

+J

Cfl
fl

to
.5

■s

cfl

03
ft

en

3

3

03

o

to

kH

o

<3

d
pq

03

fl
-H
HH
O
k->

d
•3

03
g.
^^ CO

fl ft
"S 3

fl o

■a

O

d

O

f*v

fl

3
ft

+^
eo

fl-

CS
60

W

o

3

£

1

03

"3 -a

03

3

03

■fl

U

cfl

H

cfl

O^

S^

03

03

03

t;

3,

U

l-r

OS
rH

pq oj

43

*P

o

eo

«

s

S3

rH

iH

rH

rH

co 3J

CO

CO

CO

CO

^. ^a

fl

cd

13

1)

a

l-l

<

V

xi

-t-J

d

■>

a

73

cj

u

a

cd

"3

<cd

<n

•cj

ft

en

>-,

en

a?

■d
cd

>>

>
a

ft

3
o
cj

3

o

Xl

o

CO

cd

3

t-f

cd

cj

'3

<cd
en

JO'S

<cd

M CO

r-l

c>i .Sf

CD

CD X

13
o

In

XI
+•

el

CO

1

en
<co
W3

a

cd ej

<cd
©J <cd OJ

ho

cd

tj

CO

J*

cj

•a

cd

E
3

CO
XI-

3

3

CO

j2 .3

S

o

<cd

H-
i

<3

<
X)

c

CO

XI
_cd

CO

xi

O <d
XI H "

13.3

«o

OS ft

i-< ft

CD 3

n-. to

■3 3

CO O

S»5

a is

.3 .ft

CD ps

XI fl

'? X!

^ J7

CO

rl

3 ««

CO

5"

CJ

bo

. CO

r-I (M

ce ce

CJ cj

X >->

CO .

13 <cd

O CJ 1

■9 3

Xi S3
£ &

* TV

-. co

X T3
cd 13

X °
■* t!

1-3

^ cO

O gp

J*
■0 o

■d u
ft'?

* a

&

ft

3

cd

T3

3

CO

cd cj

<« p,

XI *

CO «>

&■ ^

3 cd

ax

W 3 .3
S; p cj

•d
3
X
P-

°
\

m

CM

to

cd

cd

T3

3

cd

X

o "d
^ <u
G i-i
cuo

£3

3

cd +3

a 8

CJ

co cd

ft
cu cd

3 ..

cd *—

-d ^

« x

+3 T)
.. *>

•3 ^

CO bC

g< 3
XI S"

1 £
o ^

■I""'

cd in

<u cd

3
3

he
3

Q

X CJ

cj t3

o
XI

n

cd

. -a

2 w

<s

« 3

o o

u b

ho O

" M
B c
.5 '-' ,
C ij ■
if
ft<c3

3 s

cu XI

3

o

s

CS

>

CJ

3
cd

M '3

_ S 'to

si^

CD +i V3

pi .-a

3 -d

" ft

cd cj

"3 >

cd p
4> <cd
cu (-+

cd
o
O

3

hX)

3

ft

cj

3
w O
3 "3
O CO

cfl

SxT-i

3

03 ID

ft<3

a-s

o

09

co Xi

cd cd
cj CJ
ftT3

& "

™ ft
-i ^
XJ cd

a a
I B
l|

U5 3

.- cd

CJ *S

15

toM

cd

cd cj
o ><

O g

3
o
o

cd
ccf

en

cd
ft<-

>><§
3 co
cd <cd

a M
a §

cd

I:

bo

P -d

3

ft o

"3 3

is- 5
cj ja

r<S

en S

a' 3
a s

h-3 ejP*

S.3 8 «

190 THE ISLAMIC INFUSION

trade with Syria (and therefore with the Yemen as well) ; discredit them among
the Bedouin, on whose respect their whole system depended; and make their
position in barren Mecca untenable. The following year they organized a major
campaign against Medina. They collected all available Meccan resources for
the expedition. Arrived at the Medina oasis, they set about cutting the new
grain, standing in the ear, and so forced the Medinese to abandon the strong-
holds to which they were inclined to retire and to come out to a pitched battle.
Muhammad took a strong position against a hill, Uhud, at the northern side
of the oasis. (Some of the 'waverers' preferred to sit the battle out, even so.)
Muslim tradition has it that what promised to be a victory was changed to
defeat because some of Muhammad's men, posted to guard the flank, broke
away against his orders to join in the plundering. A Meccan cavalry captain
called Khalid saw the opening and turned the tide. Muhammad himself was
wounded, but he held his ground and became the centre of a rally at the hill.

The Quraysh were overjoyed at their victory. Their women had followed
the army to encourage it, as was common on major occasions, and they cele-
brated the victory after their fashion. Indulging in unusual excess, Hind, wife
of the leader, Abu-Sufyan, tore the liver from the body of the fallen Hamzah,
Muhammad's uncle and an early convert, and bit into it; for Hamzah, one of
the heroes at Badr, had there killed her father.

With the Muslim army still partially intact, however, and with some forces
in Medina not yet engaged, the Quraysh evidently did not feel strong enough
to attack the Medinese strongholds; they withdrew with some restoration of
prestige but without subduing Muhammad. 1 Muhammad took advantage of
their departure to exile a second Jewish clan, the Banu Nadir, whom he sus-
pected of hostile designs; when they refused to leave on the same terms as the
Banu Qaynuqa', retaining ownership of their palm groves, he besieged their
strongholds and forced them to leave and forfeit their palm groves also. Out-
side Medina, Muhammad assured himself of the co-operation or at least
neutrality of any tribal groups who stood to gain by friendliness with Medina,
and he continued his raiding.

Two years later, when it became clear that Muhammad was becoming
stronger rather than weaker, the Quraysh made a still greater effort. They
summoned all the Bedouin allies still left them, adding that strength to the
full local strength of Mecca. They thus admitted that Meccan strength by it-
self could not put down Muhammad. The campaign had to be decisive, or even
the potential ultimate strength of the Quraysh would prove insufficient and
prestige might be irrecoverably lost. The cumbrous coalition this time arrived

1 Frants Buhl, Muhammeds Liv (Copenhagen, 1903), p. 251 (p. 256 of the German
translation), suggests that the failure of the Quraysh to follow up Uhud resulted from
their lack of statesmanly vision, which let them be satisfied with formal vengeance for
Badr. The case is conjectural, but in view of the leadership which some of the same
Quraysh later took in major Muslim conquests, W. Montgomery Watt's analysis of
the event followed here, is more convincing. Similar dilemmas arise in interpreting
many events of Muhammad's life.

THE EARLY MUSLIM STATE I9I

later, after harvest, and the Medinese could not be lured from the more built-
up part of the oasis and their strongholds there. To neutralize the Meccan and
Bedouin cavalry {the farmers of Medina had few horses), Muhammad had a
ditch dug across the more vulnerable sectors. For about a month the ditch
was successfully defended in a series of skirmishes and the fighting was re-
stricted to what could be done on foot. Then some of the Bedouin were per-
suaded to abandon the Quraysh, and the whole company faded away. Muham-
mad's blockade of Meccan trade was confirmed. From this point on, the
Quraysh had defensively to await Muhammad's moves.

The one Jewish clan in Medina that still resisted Islam and Muhammad's
leadership, the Banu Qurayzah, had remained neutral during the defence of
the ditch but had negotiated with the Quraysh. The exiled Jewish clans had
been very active in supporting the Bedouin coalition in favour of the Quraysh.
When the Quraysh departed, Muhammad attacked the Banu Qurayzah, re-
fused to allow them to depart into exile like the Banu Nadir, and insisted on
unconditional surrender. In Arab expectations (as among many ancient peop-
les), when enemy captives were taken, the women and children were enslaved
but adult males were killed or held for ransom, as they were not dependable
as slaves. Muhammad now allowed no ransom but insisted that all the men,
about six hundred, be killed.

Muhammad had been able to set up a new moral order in Medina and had
been able to defend it against the Quraysh attacks. But even so it was not self-
sufficient. In a society where the Bedouin set the moral norms of all, a single
oasis could not long maintain, by itself, quite contrary standards. More speci-
fically, Mecca might for now be neutralized; but so long as the Meccan system
still stood on its pagan foundations, any other system in the area was pre-
carious. If Muhammad was to create a totally responsible moral environment
for his Muslims, he must Islamize the Meccan system itself. Consciously or not,
what he proceeded to do would be sure to have that result — and more.

Muhammad's first public preaching had taken place in the midst of a war
between the Sasanian empire and the Roman — a war in which at that time
Syria was being occupied by the Sasanians. The occupation was a major cala-
mity for the Byzantines: not only was the territory devastated by the armies
as they occupied it, but large numbers of the most crucial personnel were de-
ported to old Sasanian territory — for instance, many of the inhabitants of
Jerusalem. Even beyond Syria, Constantinople itself was besieged. For the
Quraysh, the occupation and devastation put in question their trade, of course,
but also their neutrality. What had been a strong but distant empire had now
moved closer — both in the Yemen and in Syria. What Muhammad had in
mind we cannot know, but if all his expeditions that now followed had been as
successful as were those against the Quraysh, he would have been gathering
into his system not only the elements which had gone to make up the Meccan
system, but also the Arabs at the northern and southern ends of the main
Meccan trade route — taking them away from either Abyssinian or Sasanian or

192 THE ISLAMIC INFUSION

Byzantine control and aligning them with the trading cities of the Hijaz. If
this alignment had in fact succeeded, it would have created a power from
Syria to the Yemen which might conceivably have defied both Byzantines and
Sasanians, even if the latter continued to hold the main part of Syria.

From his base in Medina, and presumably building upon the ties of alliance
that already existed between various Medinese clans and some of the neigh-
bouring Bedouin tribes, Muhammad had been systematically building up
Bedouin connections to rival those of the Quraysh. If not a full allegiance to
Islam, the Bedouin accepted at least alliance with the Muslims and some re-
cognition of Muhammad's leadership. This was accomplished partly by direct
conversion; partly, when some anti-Muslim act gave occasion, by punitive
raiding (in parties ranging from a dozen to several hundred). But mostly it was
by diplomacy. Muhammad played on differences between factions in a clan or
a tribe as well as offering the more direct advantages that might come from
the friendship and perhaps the arbitration of a neutral Medina and, later,
from sharing in the Muslim raids. Thus the system of security among clans,
which he had built in Medina, was extended into considerable territory be-
yond the oasis.

The setback at Uhud had not long interrupted this policy. Gradually it be-
came clear that Muhammad was aiming at converting all accessible Arabs to
his faith, not merely those who had had links to the Quraysh; or at least at
tying them to his society by bonds that would assure peace and security a-
mong all the Arabs, such as would be congenial to the new Islamic ideals; and
most especially the Arabs to the north — toward Syria. The year after Uhud, a
Muslim expedition into the Syrian desert underlined his interest in the Syrian
Arabs. Gaining them over would mean, in the first instance, disrupting the
Meccan trade with Syria. But there was more. The tribes in that direction had
been accustomed to varying degrees of Byzantine political influence, many be-
coming more or less Christian as well. Since Syria had passed under Sa.sa.nian
administration, the Sasanian-Byzantine rivalry seemed ended in favour of the
Sasanians. Muhammad, however, did not believe the Sasanian victory was
final. In the Qur'an, their subsequent defeat was predicted at the very moment
of their victory. (In 622, in fact, Heraclius had invaded the Sasanian empire
through the Armenian highlands; by 625, the year of Uhud, he was ready to
make a full push to the heart of the empire, where, indeed, by 628 he was en-
tirely successful. He forced the Sasanians to restore the status quo ante.) But
meanwhile, Muhammad's systematic efforts in the direction of Syria suggest
that he was hoping ultimately not only to replace the Quraysh in the central
Hijaz, but also (outbidding the Sasanians) to replace Byzantium, and its Chris-
tian allegiance, among the Arabs further north.

In Medina, after the Meccan failure before the ditch, Muhammad presided
without open opposition over a society of Muslims, with a certain number of
Jews remaining in a more or less autonomous dependent relation to the various
Muslim clans. Among the tribes outside Medina, most were pagan and were

THE EARLY MUSLIM STATE I93

increasingly required to become Muslims as a condition of entering into league
with Muhammad and into his security system. But, especially on the way to
Syria, many were more or less Christian, and in the oases most were Jewish.
From these allegiances there was little conversion. Many of the oasis dwellers,
who had been shown to be incorrigibly inimical, were reduced to dependence
by military expeditions, beginning in the year following the defence of the
ditch, and were forced to turn over part of their crops henceforth to Medina.
In contrast, some Christian-oriented Bedouin tribes who were willing to
work with Muhammad were accorded, it seems, a status as equal allies.
Yet Muhammad was always acknowledged as commander of the joint
enterprises.

Accordingly, Muhammad's society came to include both Muslims and non-
Muslims in various degrees of membership. It had long since ceased to be just
a new tribe of the faithful, or even a local voluntary association. It was be-
coming a complex and extensive society of heterogeneous elements, more fully
organized than had been the Meccan system (both religiously and politically) ;
the political structure which Muhammad was building for it was by now
clearly a state, like the states in the nations round about Arabia, with an
increasingly authoritative government, which could no longer be ignored with
impunity. Muhammad sent out envoys, who taught the Qur'an and the
principles of -Islam, collected the zakat, and presumably arbitrated disputes
so as to keep the peace and prevent feuding. The Muslims of Medina thus
undertook to bring into being throughout much of the Hijaz, and even beyond
it, a way of living which should be just and godly. They depended- funda-
mentally, to be sure, on the willingness of a majority to accept the system for
the sake of its more immediate benefits in peace among themselves and
strength against rivals outside. But the ideal was to be established whether
with or without the active co-operation of the various tribes.

Mecca is taken

Such a system could not, however, well be completed or even survive without
including Mecca and exploiting the trade route between Syria and the Yemen:
without actually replacing the Meccan system. In 628, toward the end of his
sixth year in Medina, Muhammad marched with perhaps a thousand or more
men to Mecca with the stated aim of taking part in the annual hajj pilgrimage.
After tense negotiations at Hudaybiyah outside Mecca, he signed a treaty
with the Quraysh: the Muslims would withdraw this year (it took all Muham-
mad's charisma to hold his men to this renunciation), but the following year
the Quraysh would evacuate their own city long enough to allow the Muslims
to make the pilgrimage without hostile contact. Temporarily, at least, the
Muslims would control even the Meccan shrines. The treaty could seem highly
favourable to the Quraysh: Muhammad allowed them a ten-year truce, during
which their trade would be unhindered. But the Quraysh had to permit their

194 THE ISLAMIC INFUSION

Bedouin allies to leave them and join Muhammad — which, indeed, some of
them promptly did. Given the atmosphere of high prestige that Muhammad
already had, his tribal system in the Hijaz was, in effect, tacitly being accorded
Quraysh approval; and the old Meccan system was being allowed to lapse. Yet
in the negotiations Muhammad had demonstrated a friendly, even generous
attitude to Mecca. The Quraysh could be assured that they would hold a high
position in his system if they should enter it.

In the year following Hudaybiyah, Muhammad completed his subjugation
of certain major Hijaz oases. The pilgrimage was duly made as arranged. On
the reoccupation of Syria by the Byzantines in 629, he sent a major expedition
(3000 men) to Mu'tah at the southern tip of Syria, which made a show of
force.

Then in 630, in Muhammad's eighth year at Medina, Muhammad inter-
preted a skirmish between some Bedouin allies of the Quraysh and of the
Muslims as a breach of the treaty by the Quraysh. Collecting all his Bedouin
allies, he marched to Mecca with an enormousl}- increased host — some ten
thousand men. After the death of many leading men at Badr, Abu-Sufyan (of
the Umayyad clan) had become the most prominent leader among the Quraysh.
Since the breach of the truce, he had evidently been attempting to arrange
some settlement by personal negotiation; now he came to Muhammad's camp,
reluctantly became a Muslim, and returned to Mecca announcing that Mu-
hammad would grant a general amnesty if he were permitted to enter the
town as master. The Quraysh agreed. There was little resistance when
Muhammad marched in. He received the peaceful submission of almost all his
old enemies; a handful were proscribed, chiefly for public insults against him
in verse.

The Muslims had now inherited the position in Arabia of the Quraysh. The
Quraysh immediately joined Muhammad in an expedition against those Bed-
ouin who still resisted, and they were so generously rewarded out of the spoils
that Muhammad's older followers were inclined to complain. Most of the
Bedouin were forced to submit very soon, and submission now meant full ac-
ceptance of Islam. The idols, sacred stones, and shrines of godlings in Mecca
and in all the areas dependent on Mecca were destroyed. Before long, Ta'if,
Mecca's rival and partner as a trading city, which had at first successfully
resisted a siege, found itself isolated and forced to submit likewise.

The year after the taking of Mecca is known as the year of deputations.
Representatives from tribes all over the Hijaz and Najd arrived to come
to some understanding with the new power. In some cases, whole tribes
were ready to adopt Islam. Often just one faction within a tribe seems to have
come seeking support against its rivals. There were a few deputations from al-
most every part of Arabia, even areas remote from the Hijaz such as the
Bahrayn mainland and 'Uman. An important part of the Yemen, where Sasa-
nian control seems to have become weak during the wars, submitted, notably
the Christian town Najran; as monotheists, the people of Najran were

THE EARLY MUSLIM STATE 10.5

permitted to acknowledge the political control of Muhammad without
abandoning religious allegiance to their own prophet, Jesus.

The tribes toward Syria in the northwest, however, were mostly still un-
ready to submit; after the Sasanian defeat, they seem to have renewed their
ties with Byzantium. Muhammad seems to have been more concerned about
them than about any of the others and his remaining military efforts were
mostly directed against them. From the time of the Sasanian withdrawal,
those who submitted to Muhammad were given very good terms, while those
who refused had heavy tribute laid on them when they were overwhelmed. A
year after taking Mecca, Muhammad led his largest expedition, perhaps
30,000 men, against the Band Ghassan, the chief defenders of Byzantine inter-
ests on the Syrian frontier, with indecisive results.

At the next hajj pilgrimage to Mecca, those Arabs who remained pagan
were forbidden to appear thenceforth. Then in 632 Muhammad made the pil-
grimage in person, establishing the forms of the pilgrimage which were to hold
in Islam. In the Islamic system, as in the Meccan system, the pilgrimage had a
prominent place; but in being Islamized, its cult was more sharply focused.
It was intertribal no longer because it assembled tokens from all the tribes but
because its cult far transcended any tribe — even the Quraysh. Though obser-
vances at several secondary places were retained, these were all made to de-
pend upon the primary visit to the Ka'bah, the house of Allah founded by
Abraham and Ishmael. The Muslims kissed the Black Stone in the corner of
the Ka'bah no longer as embodying some godling, but as a symbolic act of
allegiance to God, who had sent both Abraham and Muhammad to guide man-
kind.

A few months later, in the midst of equipping another expedition toward
Syria, Muhammad was taken with illness and died in the arms of 'A'ishah. He
was buried at the spot where he had died.

The genesis of a new regional culture

The period that followed was naturally of supreme importance in forming
the Islamicate civilization. But our interest in it here is for the elements going
to form a civilization which in itself did not exist till later, rather than for the
general cultural life of the time in itself. Throughout the period of genesis, be-
fore and for a time after Muhammad, the mainstream of religious, artistic,
intellectual, and commercial life in the region from Nile to Oxus continued to
reflect the ascendancy of earlier cultural allegiances. In the light of Islamicate
cultural developments, it is only a limited range of what was happening, in all
this period, which stands out as specially pregnant for us. Within this range
we often include events equally significant for their own cultural setting as for
Islamicate development: thus certain evolutions in the pre-Islamic Roman and
Sasanian empires, in whose territories the Islamicate civilization developed,
were decisive under Islam also. But sometimes, though we exclude much that

I96 THE ISLAMIC INFUSION

loomed large at the time, we include events which at the time possessed little
significance for the dominant cultural life of the age. For instance, the culture
of pre-Islamic Arabia, also crucial later under Islam, was in its own time mar-
ginal; to the rise of an artful but not highly rich or diversified tradition of
Bedouin poetry, neither Greeks nor Aramaeans nor Persians of the time
would have had reason to pay much attention.

Even the emergence of Islamic faith and the expansion of Arab control,
which introduced ultimately decisive new traditions into the whole cultural
situation, had at the time a limited impact and might have seemed transient
and superficial. From the perspective of the older civilizations, the minority
community of Muslims did not then represent, despite their power, the highest
or most significant levels of culture; it is only in retrospect that they become a
major focus of our attention.

This minority group presents one point of overriding interest. They repre-
sented consciously and intentionally a new tradition set over against the great
traditions of ancient civilization. Gradually some of them began to imagine
replacing the former societies of all mankind with a new society based on their
new ideals. At length, we have something approaching a total social experi-
ment — perhaps one of the few really major ones in history. To be sure, this
did not begin to mature till the very end of the period; with its maturing, we
enter the realm of Islamicate civilization proper. But its seeds were present in
the small group of primitive Muslims.

In the formation of the civilization, therefore, a small group of concerned
Arab families, with more or less Arabian habits of thinking, were peculiarly
important. What the Irano-Semitic traditions of the settled lands did to modi-
fy their expectations and life-patterns, against their pre-Islamic Arabic back-
ground, had lasting effects among Muslims generally. At the same time it must
not be forgotten — as too often it is — that from a broader historical viewpoint
the Arabs were essentially foreigners being assimilated into an ongoing cultural
pattern, which they helped to modify, partly through some bits and pieces of
their own older heritage, but chiefly through two things equally new to all
concerned: their catalytic presence as a new ruling class; and Islam itself, of
which they were the carriers. What we are dealing with here is the history of
the whole Irano-Semitic historical complex and its transformations; we deal
with the internal history of the new ruling class only as it is relevant thereto. 2

2 Failure to recognize the minority role of the Arabian families in the social develop-
ment of the time has resulted in one of the most distortive tendencies in all Islamics
studies, what may be called the Arabistic bias, which has pervasively twisted the analysis
not only of the early period but of all Islamicate history. Among other results is a recur-
rent notion that Islamicate civilization was 'sudden' in appearance and flowering — as if
it had no direct background save the Arabian desert, and all else were 'borrowed' and
quickly incorporated by nomads. Compare a more explicit analysis of the bias, toward the
end of the section on the history of Islamics studies in the Introduction under 'Historical
Method'. See also my 'Unity of Later Islamic History', Journal of World History, 5,
(1960), especially pp. 880-82.

THE EARLY MUSLIM STATE I97

The establishment of the caliphal state

In pursuit of a new and total moral order, Muhammad had reconstituted most
of the elements of the Meccan system of the Quraysh in a new, broader system,
which, however, maintained and even extended the neutral independence of
the Quraysh on both the political and the religious levels. But this social labour
had been largely personal to himself. The Qur'an, as such, had supported this
side of his work, but its emphasis was on the more individual level. It had,
typically, provided for no political contingencies on the Prophet's death.

The first question that arose on Muhammad's death was whether any state
should survive it at all. Islam was a personal relation of men and women to
God. It had been preached by a prophet, indeed, and so long as he lived it
could be presumed he would provide the safest guidance to God's will. On his
death, each group of men that had accepted Islam could be expected to find its
own way to obey God; unless, indeed, God sent other prophets to be followed,
as might well be anticipated. The Qur'an referred to numberless prophets and
gave no clear indication that Muhammad was to be the last of them. Indeed,
more than one monotheistic prophet had actually appeared in recent years, in
tribes beyond Muhammad's main sphere of action, presumably inspired by
Muhammad's example. The most prominent of these was Maslamah (called in
scorn 'Musaylimah'), among the Banu Hanifah in central Arabia; if Muham-
mad had denounced him in his lifetime, Maslamah's followers could suppose it
was because Muhammad was jealous that revelation should come to any other
than himself; after Muhammad's death, they could have little doubt of the
part of one who would be faithful to God. Others also might think to turn to
him.

Of the Bedouin tribes that had submitted to Muhammad, many felt them-
selves free of any further obligation and, with or without new prophets to turn
to, refused to send any further zakat to Medina; many others seem to have
waited only to see what Muslims at Medina and Mecca would do, for the power
of the Quraysh, at least, had not been negligible even apart from Muhammad.
At Medina itself there was consternation and indecision. The Ansar, the Mus-
lims of Medina, were soon suggesting that they should choose a leader for the
Muslims of Medina and the Quraysh should choose a leader for those of
Mecca.

Such were the most obvious resolutions oi the crisis. But some men had a
more ambitious conception of Islam and of the Ummah community Muham-
mad had created. Islam was not merely a matter of each individual's obeying
God; it was a compact in which all Muslims were bound to each other as well.
This compact did not cease with the Prophet's death; the pattern of life he had
instituted could be continued under the guidance of those who had been closest
to him, the earliest Muslims. Any who separated from the core of the Muslims
at Medina were in fact backing out of Islam itself; they were traitors to the
cause of God for which Muhammad and his followers had so long been fighting.

198 THE ISLAMIC INFUSION

That cause was still to be fought for, and demanded a single chief to whom all
would be loyal.

Abu-Bakr and 'Umar are credited with persuading the Muslims of Medina
to this audacious viewpoint. They broke in among the assembled Ansar leaders
and called for unity: 'Umar promised his loyalty to Abu-Bakr, and the Ansar
soon followed suit, as did the Quraysh. With the two towns thus determined
to maintain Muhammad's polity, the demand was extended to the Bedouin.
To subdue them, Muslim energies were thrown into the wars of the Riddah, of
the Apostasy', as it came to be called on the ground that the recusants had
apostatized. The Muslims found a general of genius in a latter-day convert
from the Quraysh, Khalid b. al-Walid, who had distinguished himself as an
enemy at the battle of Uhud. Recalcitrant tribes were attacked in several
quick campaigns and reduced to obedience. But Muhammad's system had
been reaching out to clans and factions in tribes much further afield. Such
factions could not remain as they were; they had either to be vindicated or
left in the lurch. In the snowballing impetus of enthusiasm, many tribes in
which only a minority faction had recognized Muhammad — and even some
where none at all had done so — were now forced to acknowledge Islam and pay
zakat to the collectors from Medina. The several new prophets that had arisen
were declared false — from this time on it was held that there could be no
prophet after Muhammad, a doctrine then equivalent to asserting the unity of
Muslims. It was the community organized at Medina that was to be the
acknowledged authority in Islam, and it was Muhammad's associates there
whose decisions were to be accepted in all matters of common concern. At the
same time, all the Arabs, as Muslims, were to be essentially equal under the
terms of that leadership. Within less than two years the power of the Muslim
community, so reconstituted, was far more widespread than it had been under
Muhammad.

Thus were ruled out two possible outcomes of Muhammad's challenge. One
could have imagined a continuing wave of prophetic leadership, inspired men
in the several tribes carrying on Muhammad's tradition in a Bedouin Arabia
fragmented but confirming its cultural independence. Or one could have im-
agined an assimilation of Muhammad's mission to the Hebrew Bible, which
would have led eventually to its submersion in a wider Jewish tradition. The
affirmation of Muhammad's political construction meant instead that the
Arabs would be both united and independent. But this unity could prove
feasible only through carrying conquest into the lands about, for which central
leadership was required.

Before the campaigns to subdue the Bedouin were completed, some of the
Arabs were already launching raids against the Sasanian and Byzantine em-
pires. Muhammad himself had been planning a major expedition toward Syria
at the time of his death, which was duly sent ahead. Though it withdrew after
making a demonstration, it was followed up late in 633 with several smaller
raiding bands into southern Palestine. The frontier Arabs of Syria were no

THE EARLY MUSLIM STATE I99

longer being subsidized by the Byzantines, whose funds were exhausted; they
put up little resistance. In contrast to the Byzantine penury, it was excessive
bureaucratic prosperity that had tempted the Sasanians to weaken their Arab
ties by abolishing the subsidized Lakhmid Arab kingdom on their frontier and
undertaking to control their dependent Arabs directly. Some independent
northeastern Arabs had already (notably in 610) had some success against the
Sasanians; they likewise now launched a raid on the frontier of the Iraq. They
now had the co-operation and, as operations increased in scale, accepted the
leadership of the Hijaz Muslims. Under the bold and far-sighted generalship of
Khalid b. al-Walid, the raids against the exhausted empires proved successful
and yielded much booty. In particular, Hirah, the former Lakhmid capital,
was occupied. Various tribal groups came together to share in the work. The
leadership of the Medina Muslims was the only common arbiter making pos-
sible the large-scale co-operation necessary; the participating Arabs accepted
it and called themselves Muslims. As soon as the raids northward were well
advanced, there was no longer any question of pagan Arabian tribes refusing
to acknowledge Islam.

In the following years, the leadership at Medina had two cares: to spread a
more serious Islam among the tribes and to organize the raids on the empires
into expressions of Muslim power. For the first purpose, Qur'an-reciters were
sent, as they had been under Muhammad, to teach the Arabs the essentials of
Islamic faith. (Tribes already Christian, however, were not expected to be-
come Muslim.) But such teaching merged with the second objective: to or-
ganize the Arabs and lead them on campaigns. The moral and financial soli-
darity implied in the Qur'anic teaching became the foundation of the military
expansion. In carrying out the second objective, the Medina Muslims made a
further major decision by 635. From raids for booty or for, at most, a border
lordship over the nearby peasantry, the campaigns were extended into a full-
scale conquest of the settled lands. Henceforth the Muslims aimed at occupying
their cities and replacing their governments with Muslim government.

There was no attempt at converting the peoples of the imperial territories,
who practically all adhered to some form of confessional religion already. Islam
was felt to be primarily, if not exclusively, meant for Arabs, and only within
the Peninsula was there any sense that all ought to be Muslims. Yet even
Christian Arab tribes were still allowed to participate actively in the conquests.
In the chiefly non-Arab agricultural lands, .the object was not conversion but
rule. The limited example of Muhammad in subjecting settled Jews and Chris-
tians in western Arabia was extended beyond Arabia to all lands within reach.
The superiority of Islam as religion, and therefore in providing for social order,
would justify Muslim rule: would justify the simple, fair-dealing Muslims in
replacing the privileged and oppressive representatives of the older, corrupted
allegiances. The caliphal state was no longer simply an Arabian common-
wealth but was a vehicle of conquest beyond Bedouin Arabia, and depended
on that conquest for its financial and psychological existence.

200

THE ISLAMIC INFUSION

Chronology: AM-Bakr to 'Abd-al-Malik, 632-692

632-656 Military occupation of the empires, directed from Medina

632-634 Caliphate of Abu-Bakr: Arab tribes are defeated in the Riddah

wars, establishing the leadership of Medina in a single Muslim
society, in which all Arabia is incorporated

634-644 Caliphate of 'Umar: Most of the Fertile Crescent, Egypt, and

much of Iran are conquered, and the patterns of military
settlement and of finance of the Islamic regime are set up

644-656 Caliphate of 'Uthman: Conquests continue northward, eastward

in Iran, and westward from Egypt, but with the enormous
enrichment of privileged families at Medina and Mecca,
jealousies and discontents divide the Muslims; the Qur'an text
is standardized for the sake of unity

656-661 The first fitnah: 'Uthman is murdered, and in civil wars 'Ali,

established at Kufah and at first widely recognized as caliph,
gradually loses Muslim allegiance; Medina is abandoned as
capital, and factions are organized, especially that of the
intransigently puritan Kharijis (658)

661-683 The Sufyani Umayyads:

661-680 The Umayyad Mu'awiyah is caliph at Damascus, relying on

Syrian Arab power and Muslim desire for unity; conquest is
resumed, especially in the Mediterranean (with a powerful fleet) ;
internal discontents are restrained by threat of force; Ziyad b.
Abih governs former Sasanian areas

680-683 Mu'awiyah's son Yazid succeeds (idea of hereditary rule is

implied) and 'Alt's son, Muhammad's grandson, Husayn is killed
at Karbala', in an attempted rising of Kufah in the Iraq — his
death becomes symbol for partisans of 'Alid rule

683-692 The second fitnah: On Yazid's death, Ibn-Zubayr restores

Medina as capital; but at Marj Rahit (684) the Umayyads
regain Syria under Marwan, and Mukhtar al-Thaqaf I at Kufah
(685-687) tries to establish the rule of the family of 'All; under
Marwan's son 'Abd-al-Malik (685-705), in civil wars, the
Umayyads regain control of all Islamic provinces

The conquest of the Sasanian empire

In 634, two years after Muhammad, his lieutenant Abu-Bakr had died, leaving
'Umar as his acknowledged successor. 'Umar may have been responsible for the
decision to occupy the agricultural provinces; in any case, he carried it out
systematically. Some initial successes in 635, especially in Syria where even
Damascus was occupied for a time, may have been due to the force of surprise.
In 636 the Roman army in Syria — not the main army of the empire, of course
— was destroyed at a point well chosen by the Muslims on the Yarmuk river;
the Arab auxiliaries, forming a major portion of the Roman army, having
gone over to the Muslims at a crucial point. Most of the Syrian cities then
capitulated with little struggle. This encouraged the Muslims to make a more

THE EARLY MUSLIM STATE 201

concerted and highly organized effort against the Iraq. In 637 the main army
of the Sasanians was destroyed at Qadisiyyah, guarding the Euphrates. Pre-
sumably it was after this battle that the Sasanian Arab auxiliaries went over
to the Muslims. Most of the cities of the Iraq then capitulated. In the Iraq,
among the cities surrendered was the capital of the empire, Ctesiphon, where
little resistance was offered. By 641, when the Roman emperor Heraclius died,
practically all of the Aramaic-speaking lowlands had been occupied, including
the Jazirah (Mesopotamia proper) in the north and the Karun (Dujayl) valley
in Khuzistan.

The Roman provincial power in Syria and the central Sasanian power in
the Iraq seem to have lost all morale and to have collapsed without serious
attempts at internal co-operation or regrouping. In Syria, at least, this appears
to have resulted from the apathy not only of the peasants but even of the
urban populations, who in the Agrarian Age participated somewhat in the
privileges of rule and would normally obstruct usurpation by an alien group.
The large body of Syrian Jews had long been persecuted by the Christian em-
pire, and had actively assisted the Sasanian conquest; but they had no more
reason to trust the Muslims than the Byzantines. But the majority of Syrian
Christians were likewise persecuted, for they refused to accept the Greek
church leadership which held power at Constantinople, and the creed of the
Council of Chalcedon which that leadership wanted to enforce. The Syrians
mostly preferred to support a religious community of their own with Aramaic
leaders, who adopted a Monophysite creed and seem to have been supported
by the Sasanians. Even an attempt at a neutral creed under Heraclius, when
he reoccupied the land, had issued only in more persecution. At the same time,
the imperial (Greek) church, which had allowed some of its treasure to be used
during the former war, demanded to be repaid and forced an unusually severe
rate of taxation. When the imperial army was broken, the city populations
accepted individual treaties with the Muslims (stipulating a lower rate of
taxation) and received them in friendliness. The Greek-speaking landlord class
withdrew to the Anatolian highlands and never came back.

In the long run, the collapse of the Sasanians in the Iraq was still more de-
cisive. Especially since Nushirvan, a large part of the Mesopotamian alluvial
plain had become virtually a state farm, in the 'SawaxT, where the revenues
were not allowed to be diverted to private landholders. It was maintained by
a massive irrigation system which was no longer manageable on a piecemeal
basis as irrigation there once had been and still was, to some degree, even in
earlier Sasanian times. This had formed the physical basis for the centralized
army and hence for the centralized, bureaucratic empire. Now, only close and
continuous central administration could keep it in order at all. A central col-
lapse would mean ruin throughout large agricultural tracts — and must con-
firm the disaster to the central treasury. But following the Sasanian defeat in
the last war with Rome, several years had passed in political chaos as various
claimants to the throne and factions in the army fought for the prize of power;

202

THE ISLAMIC INFUSION

different Sasanian provinces were run almost in independence by the generals.
A major shift in the Tigris bed seems to have created permanent swamps in
the lower Iraq and ruined much farmland there even before the war was con-
cluded. It is just possible that the changes in land formation were already

sea

Bay$an,

Conquest of the Fertile Crescent and adjacent lands to 656

beginning which eventually made much of the alluvial plain inherently harder
to irrigate than it was earlier. But the political disruption alone was enough
to account for unprecedented disaster conditions in the Iraq. This disaster, in
turn, made it more difficult for the winner in the civil wars (Yazdagird III, in

THE EARLY MUSLIM STATE 203

632) to impose his authority. Nushirvan's policies had ended, at least for the
moment, in collapse. 3

Even apart from the disorders, the population of the plain could have little
interest in holding the state lands for one government rather than another.
The majority of people there were Christian and Jewish, or Manichean, and
had suffered disabilities under the Mazdean hierarchy. The Sasanian upper
agrarian classes did not have a personal interest — or following — in the plain;
they seem to have been based chiefly on the Iranian highlands. The most im-
portant part of the Sasanian soldiery with a local interest in the Iraq were
Arabs — some of them having been lately deprived of their autonomy. When
the Sasanian forces made a strategic withdrawal to the highlands, then, the
Sasanian nobility went with them; the Sasanian Arab soldiery went over to
the independent Arabs; and neither the peasantry nor the urban populations
that remained offered resistance to the new military masters.

When Arab possession of the Fertile Crescent had been assured, a wholesale
migration of tribes from all parts of Arabia poured in, men bringing their
families along, to join the victorious armies. These soon provided enormous
army potential. The families were lodged in military bases quickly built on the
edge of the desert, and armies were sent into all neighbouring lands. The first
expedition beyond the Aramaic lowlands began late in 639 into Egypt, well
known to the Meccans for its wealth. In Egypt, the Coptic inhabitants had
likewise resisted the Greek imperial church in the name of a Monophysite
Christian creed and had been most bitterly persecuted since the evacuation of
the Sasanians. During 641, most of the country was occupied, and in 642 even
Alexandria, the local Roman capital. Within a few years, Egypt was supplying
the Hijaz with tribute grain as it had been supplying Byzantium.

In 641 began the advance into the Roman and Sasanian highlands. Despite
some initial reverses, Mu'awiyah (son of Abu-Sufyan, former leader of the
Quraysh), as governor of Syria, was able in the next few years to raid not
only into Cilicia (southeast of the Taurus mountains) but far into the Anato-
lian peninsula, reaching Amorium by 646; but he was unable to occupy much
territory beyond Melitene permanently, and by 647 was reduced to destroying
fortresses in Roman territory which, for the time being, he could not expect to
hold.

But the most important advances were into Sasanian territory from Basrah
and especially Kufah. It was to these centres that the bulk of the new Arab
immigration came, swamping the older, more city-disciplined, elements that
had known Sasanian rule and Sasanian service from the time of the Lakhmids
of Hirah; whereas the corresponding older elements continued to predominate

3 Robert M. Adams, Land behind Baghdad: A History of Settlement on the Diyala Plains
(University of Chicago Press, 1965), a study of irrigation agriculture in the Diyalah plain
(just east of Baghdad) from Sumerian times to the present, provides important archeo-
logical evidence for the collapse; though nothing can be absolutely proven for particular
years, and our reconstruction of events must remain partly conjectural. The same study
is of fundamental importance for economic history throughout the Islamic period.

204 THE ISLAMIC INFUSION

in Syria. The army sent into 'Iraq 'Ajami, the main plateau area of western
Iran, overcame a major Sasanian army at Nihavand in 641; by 643 the main
cities of the province had capitulated. Deprived of their capital and of the
state income from the alluvial plain, the Sasanians seem to have been unable
to concert their forces. In contrast to the Byzantines, whose main reserves and
administrative headquarters were intact at distant Constantinople, the Sasan-
ians were thereafter reduced to piecemeal resistance on a provincial basis.
Eventually, their whole empire was overrun and the Arabs inherited their
major resources and political potentialities. In this way, the Arabs' success in
the Iraq made it possible for them to form an enduring new empire despite
their ultimate failure to overcome the Byzantines in the latters' homelands.

The Bedouin Arabs had conquered readily enough the three surrounding
countries — the Yemen, already partly subordinated in Muhammad's own
time, no longer closely held by either Sasanians or Abyssinians, and in any
case half-Arabized long since; Syria, at odds with its Byzantine rulers in an
upsurge of Semitic monotheistic communal spirit against a Hellenizing ruling
class that had long lost the creative spirit of the polis; the Iraq, where, as in
Syria, Arabs already formed a major military element and which, in any case,
no longer possessed its own independent dominant classes. Egypt also fell
readily as, like the Iraq, it had no military force of its own and was open to
control by the strongest neighbour. It can even be suggested that — without
its being planned so — many Syrian elements, and perhaps also Yemeni, co-
operated readily enough with the Islamized Meccans, with whom they had
long been in trading relations; it is as if there was set up a Yemen-Mecca-
Syria axis which was strong enough to dominate both Egypt and the Iraq
with its military and commercial power. The Syrians seem readily enough to
have co-operated with at least their immediate Meccan governors: if it was
the desertion of the only military element among the Monophysites — - the
Arab auxiliaries — that gave Syria into Arab hands, it was the almost imme-
diate co-operation of the non-Arab Syrian shipping which allowed the Arabs to
appear as a naval power almost from the start, defeating the seriously trun-
cated Byzantine naval forces. Leading Meccan merchants lost little time in
penetrating, as privileged competition, the Iraq particularly; but it is not
clear whether they did so with any active Syrian support. But conquest of the
hinterlands of these three neighbouring countries was another matter. In such
efforts, only special circumstances could allow an Arab success.

The Arabs very early sent an expedition against Abyssinia, a naval expedi-
tion which was wrecked. Wisely, they made no further attempts. The Abys-
sinians formed a reasonably strong power, still intact, for the conquest of the
Yemen had scarcely touched them; there was no reason to expect that they
would offer the Arabs even a beachhead. Even the Nile Sudan, which had not
been integrated into the Roman empire and so retained its own social order
intact, and where there was no prior Arab base, successfully resisted invasion.
As to the Byzantines, their power was based on the north Mediterranean

THE EARLY MUSLIM STATE 205

Chronology: The Arab Conquests, 632-655

632-633 Death of Muhammad brings about Riddah wars; Abu-Bakr

brings back to Muslim allegiance Arab tribes whose primary-
allegiance was to Muhammad's political leadership; battles
fought in many different parts of Arabia; fighting groups push
out of Arabia to northwest and northeast

633 Hirah, Sasanian fortified town near the Euphrates river, taken

634 Byzantine force defeated in S. Syria

635 Damascus taken, followed by some other Syrian cities

636 Battle of Yarmuk, near the Jordan river, crushes a strong
Byzantine army commanded by the emperor's brother, who is
killed; Syria thenceforth lies open; Damascus retaken

637 Battle of Qadisiyyah, near Hirah, crushes a strong Sasanian
army commanded by the principal general Rustam, who is
killed; Iraq west of the Tigris lies open; Sasanian capital of
Ctesiphon taken

638 Jerusalem taken; Basrah, Kflfah founded as garrison towns

640 Caesarea (Palestinian seaport) finally taken; no Byzantine
power remains in Syria; Egypt invaded (end of 639) ; Khuzistan
taken

641 Mosul taken; no Sasanian power remains west of the Zagros
mountains; battle of Nihavand in central Zagros opens that
area by destroying remaining Sasanian army; Babylon in Egypt
(site of later Fustat and Cairo) taken

642 Alexandria taken; Barqah (Tripolitania) raided (642-643); raids
toward Makran coast, southeast Iran (643)

645-646 Alexandria retaken by Byzantines; retaken by Muslims

c. 645 — Muslims engage fleets from Egypt and Syria; Muslim sea power

begins
c. 647 Tripolitania taken

649 Cyprus taken — first important Muslim sea operation

649-650 Persepolis taken — chief city of Fars and Zoroastrian religious

center

65 1 Yazdagird, last Sasani ruler, assassinated in Khurasan

652 Armenia mostly subjugated; Byzantine fleet repulsed off
Alexandria; Sicily pillaged; treaty made with Nubia, south of
Egypt

654 Rhodes pillaged

655 Combined Muslim fleets shatter principal Byzantine fleet off
southwest Anatolian coast; emperor in command barely escapes

206 THE ISLAMIC INFUSION

peninsulas, which were oriented to the sea and formed a viable unity without
Syria and even Egypt. To conquer the Anatolian highlands would have
required an overwhelming force ready to occupy every town without offering
any reason for the local privileged classes to prefer an alien rule; it is doubtful
if the taking of Constantinople itself would have produced results even so per-
manent as the Crusaders' taking it in 1204. The Iranian highlands, on the con-
trary, were linked closely to the Iraq. An integral part of the Arid Zone, those
inland highlands had been closely tied to the irrigated river basins on their
flanks, especially to that of the Oxus on the north and to that of the Tigris-
Euphrates on the west (the Indus basin was separated by barriers from the
main parts of the plateau). The capital of an Iranian empire had regularly
gravitated to the Mesopotamian lowlands, from Susa to Ctesiphon. The mutual
involvement of Iranian and Semitic traditions was almost a given, in the
common arid setting; it was part of this more comprehensive, persistent
pattern, that the Iranian state had come to depend for its financial integrity
on the Iraqi Sawad.

Hence, of the three empires in the midst of which the Meccans and their
allies had maintained their neutrality, it was only the Sasanian they were able
to conquer — when they combined the full force of the Qur'anic revelation with
the potentialities of the expansive camel nomadism. Having conquered that
empire, they were able to conquer also, beyond it, nearby lands which had
not, or had not lately, been subordinated to it. It was political reasons that
gave them the Sasanian empire: particularly the current Sasanian crisis, and
then the collapse of Sasanian power when bereft of its capital. But these poli-
tical reasons were expressions of the more enduring cultural reasons for the
persistent unity of the Irano-Semitic cultural territories between Nile and
Oxus, and their differentiation from the more purely Greek territories of the
peninsulas. Only in the Maghrib and Spain were the Arabs able to conquer dis-
tant areas without reference to the Sasanian power base; but there they stimu-
lated and gave guidance to a separate movement, that of the Berbers, which
had its own momentum.

'Umar's organization of conquest

In contrast to the Christian or even the Mazdean situation, no explicit dis-
tinction between religion and state could be made among the Muslims. As al-
ready in Muhammad's time, the head of the major undertakings of the Mus-
lims was at once head of the Muslim community and head of the whole society
which it controlled. But it was within the framework of the Muslim Arab com-
munity that the course of the whole society was to be worked out. As in Muham-
mad's Hijaz, non-Muslims were mostly relegated to the position of tolerated
dependents; they were left to organize their own autonomous life under the
protection and overriding control of the Muslim community. The governing
conceptions and ideals of the dominant Arab community were defined by

THE EARLY MUSLIM STATE 207

Islam. The Islamic institutions were thus at first designed simply as practical
expressions of the various aspects of Muslim Arab life. What Arabian customs
were already satisfactory were accepted — and with the conquests, the same
applied to patterns of relation between rulers and ruled in the agrarian lands.
Customs contrary to the new faith were replaced.

The central problem in Muhammad's time had been to replace a system of
feuding within a society, in favour of a common life under a single arbiter.
Under 'Umar the same problem was renewed under new circumstances — to
bring some common discipline among the rather lawless occupiers of the con-
quered territories. Muhammad's solution, to provide for a central distribution
of funds to those at a disadvantage and a central settlement of disputes by a
divinely sanctioned moral standard, was to be adapted and extended. The
problem for 'Umar, and for the Medinese whom he represented, was to define
the nature of the authority at such a centre.

Abu-Bakr had been known as the representative of Muhammad, his khalifah
(Anglicized as 'caliph'). His had been essentially an emergency status. The
term continued to be used of 'Umar, but he later chose for formal use the title
amir al-mu'minin, commander of the faithful. The only binding authority
the Arabs had recognized was that of the military commander on the march
to new pastures or at war. Though the Qur'an enforced the idea of a com-
munity, in which individual pious action was completed by joint action in the
cause of God, it provided directly for no government other than that of the
Prophet himself. The only position, therefore, that could be felt to be legiti-
mate was still that of military commander, with authority correspondingly
limited. 'Umar saw this as his position. But war was the business of the com-
munity for the present and such a position opened a wide scope of responsi-
bility.

The Muslims accepted him as commander of the community in any matter
wherein each man could not act for himself. This position of command rested
on personal prestige; and in this case, on a religious prestige. Since any group
action beyond tribal interests was a matter for religion, we may say that it
was precisely in religious matters that he was successor, 'caliph', to the Pro-
phet; certainly his decisions were to be consistent with what Muhammad had
shown of God's will. Of course, since he received no revelations from God, he
had no independent religious authority; it was only current political questions
that he was to decide on this religious basis. In any case, his authority depen-
ded on his personal closeness to the precepts and example of Muhammad and
on his being personally recognized at Medina — and on that account also by
the bulk of the Muslims at large — as in fact representing Muhammad's
way.

'Umar's position, then, religious and military as it was, was based on person-
to-person relationships, as had been Muhammad's. But with the vast increase
in the number of persons in the community, even among the ruling Arabs (to
say nothing of the subject peoples), the organization which he set up had to be

208 THE ISLAMIC INFUSION

less immediately personal. It was, in fact, an institution capable of operating
apart from the immediate intervention of any given individual. This organiza-
tion was centred in an army diwdn, a register of all the Muslims of Medina
and Mecca and of the conquering armies (and their descendants) . The booty
from the conquest was to be distributed in individual pensions to the men (and
sometimes women) listed in the diwan, according to their rank therein. Some
prominent Muslims received revenue from particular tracts, but most received
their share through the diwan system.

This system recognized that conquest was the keynote of the Muslim state,
and helped perpetuate this situation. With distribution of booty as the most
attractive physical resource of the state, it was obviously desirable to continue
the conquests; and though this may not have been 'Umar's intention, yet the
conquests did continue and no doubt helped with their heady results to make
'Umar's arrangements workable. But 'Umar also provided for the area already
conquered to continue in the condition of conquered possessions. Movable
booty had been distributed among the armies at the moment of conquest —
with the Prophetic fifth of the booty reserved to the disposition of the caliph
for the benefit of the poor and for other state concerns. But immovable booty
— the land revenues or 'taxes' — was mostly not divided up, but kept (as /ay')
as a single source of income to be doled out to the conquering Arabs and their
descendants (in principle) from the centre, to which a fifth was sent outright.
Each Arab was to receive his due as it was apportioned by headquarters
through the army diwan, though actual finances were largely handled in the
provinces.

The Arabs — even if they wanted to — were not to settle down in the old
cities as new landlords (garrisoning had been tried in Ctesiphon, the Sasanian
capital, with demoralizing effects). They and their children were to remain in
garrison towns (misr, pi. amsdr) of their own as a separate conquering class,
living on the tribute allotted by pensions. Each garrison town was situated for
maximum military effectiveness — normally near enough to the desert that a
potential retreat thither remained open. Kufah, in the region of Ctesiphon
(and not far from the old Hirah of the Lakhmid Arab kingdom), and Basrah,
between the desert and the ports of the Persian Gulf, were the two garrison
towns in the Iraq, from which expeditions were launched to the furthest points
east; Fustat (the future Old Cairo) at the head of the Nile delta was built as
the capital of Egypt and headquarters for expeditions further west. Only in
Syria, where the occupying Arabs had already formed close local ties before
the conquest, was the main centre an old city, Damascus, rather than a new
garrison town. From there, expeditions were launched northwest against the
one great remaining enemy, Byzantium.

But as Muslims, the Arabs were not merely an army of occupation. They
were also representatives of God's good order among mankind, founded on
adherence to His revelation. In each garrison town, and in each city where
Muslims settled a garrison, a mosque was built, which was at that time a

THE EARLY MUSLIM STATE 200,

simple enclosure, usually roofed over at one end, suitable for mass assemblies.
There the faithful came together to perform public worship {soldi), especially
the Friday midday worship. The pattern of the Friday service seems to have
echoed that of the Jewish and Christian services (the latter having been based
on the former) in the general order of worship; for instance, the khutbah (ser-
mon) was divided into two parts, answering, in the older rites, to a reading of
scripture first and a less sacred reading or sermon afterwards; and it preceded
the salat proper, as did scripture reading, the Eucharist. (In details, such as
that the leader should hold a staff when speaking, it even reflected old pagan
Arabian practices.) But in total effect, the services expressed the new Muslim
vision: thus the khutbah sermon focused not on ancient scripture, but on the
living Muslim community equally in each of its two parts. The Qur'an was
used extensively, but it was uttered by every individual, sometimes at individ-
ual choice. (Compare the chart of the Muslim public cult and the diagram of
the salat.) 4 The mosque was also used for any other public activity that was
called for.

Centred on its mosque and kept in order by its commander, each garrison
town formed a self-sufficient Muslim community, dominating and living from
the district under its military control; in the process, it moulded its own people
into an Islamic pattern. In each garrison was appointed a commander repre-
senting the caliph and hence charged with leading the salat worship and the
military expeditions undertaken from there, and with managing the tribute
collected. He was expected to keep the garrison in order, settling disputes
among the faithful in a spirit of justice and in accordance with the Qur'an
where that applied. 'Umar needed as administrators men who were able to
handle the Bedouin, most of whom (especially those not from a Syrian or
Hiran background) were unused to outside controls; men who at the same
time were able to see the long-term problems of finance and administration in
agrarian territory. Such he could find among the Quraysh— and among their
allies the Thaqif of Ta'if — but sometimes at the price of their showing less
than the highest Islamic morality in their personal lives.

Despite the weaknesses of some of his governors, 'Umar stressed Islam as
the basis of Arab life. Muhammad had left many questions open in the develop-
ing life of Medina. For the life of the Muslim garrison towns, full of new con-
verts and tempted by undreamed-of wealth from the tribute, 'Umar had to
establish sharp and clear standards to prevent rapid demoralization. He sent
Qur'an-reciters as missionaries to the garrison towns, but did not leave it all
to the Qur'an. He seems to have decided what the essential minimum common

4 For a description in English of the primary Islamic cult, cf. Arthur Jeffery, Islam:
Muhammad and His Religion (New York, 1958), section V, 'The Duties of Islam', consis-
ting largely of descriptive and explanatory passages from standard Muslim authors. He
translates salat (worship) as 'liturgical prayer', a common rendering. For greater detail,
see Edwin E. Calverley, Worship in Islam (2nd ed. London, 1957) — largely a translation of
a treatise by the great Muslim scholar, Ghazali.

210 THE ISLAMIC INFUSION

The Public Cult as It Had Developed by Marwani Times

Salat (five times each day : before dawn, after noon, mid-afternoon, before
sunset, mid-evening) :
summoned to by the

A dhdn = call to worship (in Arabic)
carried by the voice of the

Muezzin (mu'adhdhin) = chanter of the Adhan,
stationed in the

Minaret (manarah) = tower at the mosque ;
it is performed, however, anywhere, only after

Wudu' = ritual ablutions (i.e., washings of face, arms, and feet)
and while facing in the direction of the

Qiblah = direction of the Ka'bah, the shrine at Mecca

(i.e., for the majority of Muslims in the twentieth century, more or less
toward the west, not east)
reciting Arabic phrases, especially taken from the Qur'an, including the

Shahddah = statement of Islamic belief, and the

Takbir = 'Allahu akbar', 'God is Great';
and composed of two or more

Rak'ah = a sequence of bowing and prostrations;
or else the salat is
performed in a

Mosque (masjid) = any place set aside for salat,
in a group lined up in rows and led by an

Imam (leader), with whose performance of the salat the others keep time,
facing the

Mihrdb = niche in the wall toward the Qiblah,

Fridays (yawm al-jum'ah) :

midday salat performed by all adult males in a
Jdmi' = special mosque ('cathedral mosque') for the whole local
community,
with, following the salat, a

Khutbah = sermon (in Arabic, and later in set form) — including mention
of the name of the recognized Muslim ruler —
preached by the caliph or his governor (later substituted by the Khatib,
sermonizer) from the
Minbar = a series of steps on which to stand ('pulpit').

Yearly :

Ramaddn = ninth lunar month, month of daylight fasting,

at end of which is celebrated the
Lesser 'Id ('id al-fitr), with a special morning group salat,

followed in the twelfth month by the
Hajj = pilgrimage to Mecca in full form, with special rituals at Mecca and
in its vicinity,

at end of which is the
Greater 'Id ('id al-adha), celebrated both at Mecca and everywhere,

with a special morning group salat, and with ritual animal sacrifices.

THE EARLY MUSLIM STATE 211

ritual for all should be, where Muhammad had scarcely had time to do so by
example even if he had cared to make such a prescription: for instance, in im-
posing the hajj as an obligatory duty. According to later Muslim tradition, at
least, 'Umar tightened up the family law, insisting (in principle) on severe
punishments for adultery; forbidding an old Arabian practice of temporary
marriage (not far from prostitution) which Muhammad seems to have toler-
ated; and giving the slave concubine a more secure status if she became preg-
nant. In general, he insisted on rigorous discipline (being especially hard on
drunkenness) and discouraged by his own example and probably by other
regulations the luxury among the Arabs that was a natural consequence of the
conquered wealth. Helped by other high-minded associates of the Prophet, he
made of Islam a puritan standard for the Arab soldiery.

This religious character of the Arab community was presupposed in 'Umar's
army diwan as clearly as was its conquering character. The diwan gave a clear
social status to all the Muslim Arabs, even those defeated in the Riddah wars,
alongside the original community in Medina and Mecca; a social status, in
turn, based not on descent but on faith. The tribesmen were, by and large, set
off according to tribe; for tribes had converted as units; yet where appropriate,
each Arab could be given his place individually. The prime criterion was one
of priority in accepting Islam. This naturally, in practice, gave pre-eminence to
the Muslims of Medina, especially since time priority was assessed in broad
categories; but any tribesman who happened to have been an early convert —
or to have performed signal services — -could have due recognition, Muham-
mad's wives and family were given a special place as closest to him. Thus the
whole Arab community was sorted out according to a strictly Muslim criterion .
The state was centred in Medina and founded on the religious prestige of
Muhammad; but it included as integral members the whole ruling class of
Arabs scattered over the conquered provinces.

The spirit of the new order was symbolized in the era which 'Umar adopted:
it dated from the Hijrah of Muhammad, when he broke with the tribal past
and went to Medina to set up a new order. The very term hijrah was likewise
applied to the migration of an individual or a tribe to the new military camp-
cities: in joining the active Muslim community, each individual repeated for
himself the essential step which had launched the Muslim community as a
whole. Along with the new era, 'Umar also consecrated a lunar calendar, which
in itself implied a break with the environment; for (knowingly or not) in the
calendar he ignored the seasonal year, interpreting an ambiguous Qur'anic
decree of the last years of Muhammad's life as ruling out any accommodation
of the lunar cycle to the seasons. Hence the Islamic 'year', which is a pure
twelvemonth, has been about eleven days short of the true seasonal year and
neither the calendar year nor its festivals have coincided with the necessities
of pastoral or agricultural life — or with the course of other calendars.

212 THE ISLAMIC INFUSION

The first fitnah wars

'Umar died in 644 at about 52 years of age, leaving a panel of Medinese leaders
to choose his successor; jealous of each other, they chose the weakest among
themselves, 'Uthman b. 'Affan, the pious early convert and son-in-law of the
Prophet. Under 'Uthman the raids and conquests continued in many direc-
tions but with diminishing amounts of booty despite the increasing number of
immigrating tribesmen. The main conquests were in the Iranian highlands.
After a pause, the Sasanian home province, Fars, was occupied by 650, and
then the armies moved on into the great northeast province, Khurasan, more
or less brought to terms by 651. Westward, after it became clear that the
Byzantines were not to be dislodged from Anatolia by land expeditions alone,
under Mu'awiyah's leadership the Arabs also took to the sea. A successful raid
on Cyprus in 649 encouraged further efforts. With the aid of Syrian and Egyp-
tian naval skills, Cyprus was occupied and the Byzantine fleet — stripped of its
Syrian section — was shattered by 655. But such operations yielded less imme-
diate plunder than those of a decade before. They were suspended by a move-
ment of revolt against the caliph by the discontented Arabs themselves.

'Uthman had continued 'Umar's policies but with less skill. Under 'Uthman
the principles emerged of what may be called the 'Umayyad' caliphate (be-
cause all the effective representatives of it, starting with 'Uthman, were of the
Umayyad family). The soldier-tribesmen (muqdtilah) of the garrison towns,
under 'Umar quartered there on a war basis, were to remain there permanently
even though warfare became only episodic, living as Arabs set off from a non-
Arab population. They were to be governed by the merchant families of
Quraysh and its allies of Thaqif (of Ta'if), most notably men of the Umayyad
family, who would uphold the central power against tribalism and localism —
a situation seemingly transient under 'Umar, which now became regular policy.
And both the soldier-tribesmen and the governors were to be held in check by
a sentiment for a common Islam, as that which made one a true Arab.

'Uthman could not avoid, as had 'Umar, allowing the richest Meccan fami-
lies to go to the provinces, especially the Iraq, and make business ventures
there — to the annoyance of the less well-advantaged local Arabs. But he did
manage to reverse a tendency to allow private estates in the Sawad, the rich
irrigation land of the Iraq. He forced those who had begun to form such to
transfer their investment to the Hijaz; there the irrigation resources of the
several oases, consequently, were worked to the full for a time. This at once
reduced the threat of a simple merging of Arab culture into that of the Fertile
Crescent, and strengthened the central power physically. But it did not make
'Uthman popular with the Meccans.

After some years, complaints began to mount up. A riot had to be punished
with Arab blood in Kufah. Especially after the Iranian campaigns were more
or less completed, some garrison towns became discontented. After a time,
most Basrans were satisfied under 'Uthman's governor Ibn-'Amir, a good

THE EARLY MUSLIM STATE 213

general who made money in peacetime but encouraged others to do so too.
The Syrians were content with Mu'awiyah. But at Fustat and Kufah nothing
'Uthman could do would satisfy. A governor's drunkenness was an obvious
crime, though no worse than what 'Umar had had to tolerate. Some com-
plained of his minor regulations of the cult, which by now was assuming fixed
time-honoured forms. Essential to the role of Islam as a pledge of Arab unity
was his insistence on the use in all garrison towns of a single standardized col-
lection of the Qur'anic revelations. He caused all deviant editions to be
burned. This was accepted in most places ('Uthman's version is the present
form of the Qur'an), but it aroused resentment among the Qur'an-reciters
(especially the revered Ibn-Mas'ud at Kufah), many of whom had had their
own versions, varying in minor details. The Kufans long refused to comply.

Many began to complain of 'Uthman's tendency to nepotism, seeing in a
clique of his relatives the cause of their other grievances. Though himself one
of the first converts, 'Uthman was of the Banu Umayyah family (the Umay-
yads), most of whom, like their leader Abu-Sufyan, had opposed Muhammad
till almost the last minute. ' Umar had made extensive use of the experience and
skill of members of that family, but 'Uthman gave them and their associates
almost a monopoly of top posts, often letting himself be dominated by them.
This made him unpopular with the Ansar families of Medina.

Finally, some in the garrison towns complained of the financial system it-
self, which 'Umar had set up but which under 'Uthman displayed its weak
points. They disliked seeing the revenues of their districts controlled, as fay',
state property, from Medina (not without some traces of nepotism again) rather
than reserved directly for themselves. Some seem to have suggested that
the conquered lands, like the booty in battles, ought to have been distributed
outright among the soldiers. In any case, no part of the revenue was to be
sent to Medina. There is some indication that 'All b. Abi-Talib, Muhammad's
young cousin (and also son-in-law) who had grown up in his household, had
already opposed 'Umar's policies and opposed 'Uthman's still more. He was
known as a mighty warrior and was felt to be a spokesman for the malcontents.
He now became a symbol of the party of protest. 5

In 656 the discontent culminated. The Medinese had encouraged the pro-
vincial garrisons in resistance — notably at Kufah, where 'Uthman's governor
was finally refused outright. A group of Arab soldiers, come back from Egypt
to claim what they felt were their rights, seem to have been cozened by 'Uth-
man's associates into returning home with false assurances of redress; when
they discovered their leaders were to be executed instead, they returned mu-
tinously. After a period of general negotiation and counterplotting, in which

' A case for an early active role of protest on the part of 'AH has been made by Laura
Veccia-Vaglieri in 'Sulla origine della denominazione "Sunniti" ' Studi Orientalistici
in onore di Giorgio Levi della Vida, 2 (Rome, 1956), 473-85, an article relevant also
to the subsequent Marwani times. All of Veccia-Vaglieri's several writings on this
period are worth reading.

214 THE ISLAMIC INFUSION

the non-Umayyad leading families at Medina seem to have been largely neu-
tral, the mutineers broke into 'Uthman's house and murdered him. (His power
like that of Abu-Bakr and 'Umar had rested on pious prestige alone; he did
not even have a private bodyguard.)

Thereupon, two dozen years after Muhammad's death, began a five-year
period of fitnah, literally 'temptation' or 'trials', a time of civil war for the
control of the Muslim community and of its vast conquered territories. 'Uth-
man had had many opponents among Muhammad's associates at Medina, who
had done little to control the mutinous soldiery. They now divided over the
spoils. 6 The mutineers, and most Medinese too, acclaimed, as new caliph, 'All,
who accepted after a brief delay. Muhammad's favourite wife, 'A'ishah, with
two of his most eminent associates among the Meccan Muhajirun, thereupon
called for revenge for 'Uthman and attacked 'Ali for not punishing as mur-
derers the mutineers, now his most ardent supporters. The mutineers main-
tained that 'Uthman had been justly killed, for acting treacherously and for
not governing according to the Qur'an; hence no vengeance was to be in-
voked. 'Ali had to accept this argument. He withdrew to Kufah, where he had
partisans, and his opponents to Basrah; for all military strength was in the
provinces. Victorious in the resulting struggle, 'Ali made his capital at Kufah.
He was able to appoint his partisans governors in most provinces; but his chief
strength lay in the Iraq. Kufah was the chief Muslim post in the Sawad, that
part of the Mesopotamian alluvial plain where investment in irrigation had
reached its peak, and where the revenues had been reserved by the Sasanians
for state purposes directly. The Sawad formed perhaps the single most lucra-
tive part of the fay', therefore. 'Ali did distribute what was in the treasury to
the soldiers, but did not get round to dividing up the Sawad, if he ever intended
to.

'Ali had not been recognized in Syria, however, and Mu'awiyah b. Abi-
Sufyan, as governor there, in turn took up the call for revenge for 'Uthman,
his cousin. 'All marched toward Syria, but extensive skirmishing and negoti-
ating in 657 at Siffin, on the upper Euphrates, was inconclusive till Mu'awi-
yah's men (who, according to the Iraqis, were finally threatened with defeat)
put Qur'ans on the ends of their lances and called for arbitration according to
God's word. Many of 'All's followers approved this way of ending fighting
between Muslims and forced him to accept it. A good number of Muhammad's
leading associates were 'neutrals', refusing to take sides in quarrels among
Muslims. 'Ali was forced now to choose as his representative one of these,

6 The 'associates' of Muhammad, those Muslims who came into contact with him per-
sonally, formed, especially in later Muslim eyes, a special body of men and women, many
of whose names are known. The Arabic term is sahdbah, for which I use the term 'associ-
ates' where most writers say 'companions'. The term 'companion' implies too close a
familiarity to include, as the expression must, those who associated with Muhammad
only briefly or even transiently. With a proper warning, such as scholars usually give
somewhere, the word 'companion' is doubtless harmless, but 'associate' seems less mis-
leading, being more impersonal.

THE EARLY MUSLIM STATE 215

Abu-Musa al-Ash'ari, whom the Kufans had made their governor in defiance
of 'Uthman but who was no special friend of 'Ali.

Meanwhile, however, some of 'All's soldiers repented of having left up to
arbitration by neutrals a question — the guilt of 'Uthman — which they felt

Events of the First Fitnah, 656-661

656 Caliph 'Uthman is killed at Medina (siege of 'Uthman's house)
by mutineers from Egypt vs. defenders, of Umayyad family

Party of 'Ali (cousin and son-in- Opposed by party appealing to

law of Muhammad) raises him to vengeance for 'Uthman,

caliphate —

led by 'A'ishah (favourite wife of
supported by the mutineers, Muhammed) and Zubayr and

Medina Ans3.r, and Kufans Talhah (close associates of

Muhammad)

— 'Ali victorious at 'Battle of ('A'ishah's) Camel' near Basrah —

Party of 'All, caliph at Kufah, — resisted still by party of

recognized in most provinces Mu'awiyah (Umayyad governor

of Syria) appealing for vengeance

for 'Uthman

657 — Stalemate at battles of Siffin on the Euphrates —
('All's commander:

al-Malik al-Ashtar)

— leading to —

658 — fruitless arbitration (both 'Alt and Mu'awiyah rejected) at Adhruh in

Syrian desert —
('All's mediator: Abu-Musa (Mu'awiyah's mediator: 'Amr b.

al-Ash'ari, governor of Kufah). al-'As, conqueror of Egypt).

'All's partisans split :

Khariji party (opposed to Meanwhile,

arbitration) vs. party of 'Ali (the Mu'awiyah gains Egypt; 'Amr b.

Shi'ah). Khariji party defeated al-'As, governor there,

at the Nahrawan canal in the Iraq

660 Mu'awiyah proclaimed caliph in

Jerusalem

661 'Ali murdered by Ibn-Muljam,
a Khariji

— Hasan b. 'Ali sells his rights as caliph to Mu'awiyah —

already had been settled by Qur'anic standards. When 'Ali refused to join
them and held to the agreement to arbitrate, they left him to form their own
camp, first at Harura', near Kufah. These included some of his most pious
followers, notably many Qur'an-reciters; they accused 'Ali of compromising
with the supporters of injustice and so betraying his trust, which was to right
the wrongs committed by 'Uthman. These extremists, the Shurdt, more com-

2l6 THE ISLAMIC INFUSION

monly (though less accurately) called Kharijis ('seceders' or 'rebels'), elected
their own commander, independent of the other Muslims. Most of the first
group who seceded were wiped out by 'All's forces, but their movement spread,
inheriting the more uncompromising claims for egalitarian justice which had
arisen among the opponents of 'Uthman.

When the arbitration did take place, in 658, the position of the mutineers
was condemned, and hence implicitly 'All as well. 'All rejected the decision
(but without repenting of having awaited it, and so without reconciling the
Kharijis, who now saw him as acting purely for personal power) and tried to
march again against Syria. But his severities against the Kharijis seem to have
discredited him even at Kufah, where he often could raise no army at all to
fight. Thereafter, in years of often desultory fighting, Mu'awiyah made steady
headway, first taking over Egypt. Many Arabs remained neutral and 'All gra-
dually lost much of his following. A second arbitration, in which most leading
Muslims but not 'Ali took part, tried in vain to agree on another candidate for
the caliphate; for Mu'awiyah's followers were by now insisting that the caliph
be he, and most others were not yet ready to accept this. The arbitrations had
had little success except to discredit 'All's claims to the caliphate. But Mu'awi-
yah's forces proved able to defeat local resistance in Arabia. In 661, 'Ali was
murdered by a Khariji; his son, Hasan, was elevated by his still loyal following
at Kufah, but came to an accommodation with Mu'awiyah whereby he re-
tired in wealth to Medina. Mu'awiyah, who was a brother-in-law of the Prophet,
was then accepted in all the provinces as caliph.

During the civil strife, the Arabs had been driven out of the province of
Khurasan by upholders of Sasanian power and had made no advances else-
where; these setbacks were soon recovered. But enduring party strife had
arisen. The handful of surviving Kharijis had established a pattern: at Basrah
and Kufah, already under Mu'awiyah's caliphate, more than once zealous
groups of men set themselves off against the bulk of the Arabs in little war
bands, summoning the Muslims to higher standards and meanwhile living by
pillage and tribute. They believed themselves the only true Muslim com-
munity, the only genuine supporters of divine justice. They conceived that the
way of true Islam implied making war on all self-styled Muslims who did not
accept the Kharijis' own rigorous standards; and, leaving the garrison towns,
the more activist of them proceeded to do this whenever practicable. More
immediately important, the Arabs of the Iraq and those of Syria had become
enduringly embittered against each other. The Iraqis were quiescent for the
time, but in Kufah 'Ali and his family continued to be regarded as symbols of
local power against the Syrians. And potentially 'Ali was even more than this.
He had ceased to stand, as he may have at first, for an abstract principle; this
the Kharijis had made impossible. But his fate took on all the more meaning
on a symbolic level. He could be seen as a great and serious man (and he was
certainly intensely loved by those who did adhere to him at all costs), caught
by the ignoble logic of events and abandoned and dragged down to a defeat

THE EARLY MUSLIM STATE 217

which he may have deserved on the level of practical politics, but which his
personal stature could make seem intolerable. His figure was an appropriate
one henceforth round which to rally those who would protest against the
logic of events, as well as against the injustices of a centralizing government.
Very early his story came to be written as that of the noble man ruined by the
inconstancy of his friends as much as the malice of his enemies and finally by
that power of brute force which we must inexplicably confront.

On the other hand, there were many Muslims everywhere for whom the
fitnah wars had been a lesson in the importance of Muslim unity. For them,
the jama' ah, the whole Muslim community taken collectively, took on a special
spiritual importance as being under divine protection. But it was the neutrals,
whose allegiance to Mu'awiyah had been reluctant, for whom the principle of
the jama' ah remained primary. They did not merge with Mu'awiyah's own
Syrian partisans. They regarded the Syrian power as a stopgap, and if they
had been dissatisfied with 'All, they remained ready to condemn Mu'awiyah
also if he failed their high expectations. Mu'awiyah could re-establish 'Uth-
man's policies, but he could not regain 'Uthman's prestige. Even those who re-
jected the incipient Shi'ism of the time, in favour of loyalty to the whole com-
munity in its actual historical fate, had become potentially oppositional in
spirit. On 'Uthman's murder, the centre of power had been irretrievably with-
drawn from the ideally neutral Medina, lying in the shadow of Muhammad's
prestige, to the provincial garrison towns where lay military might. Hence-
forth the Muslims were not to be held together without calling into play the
might of one faction or another.

In launching the venture of Islam, the events of the first generation after
Muhammad were almost as formative as those of Muhammad's own time. It is
not accidental that later Muslims have identified themselves in terms of these
events and of the factions that grew out of them. They have interpreted the
whole of history in symbolism derived from them, and have made the inter-
pretation of those events and of the leading personalities in them the very
test of religious allegiance. This has confused the factual historical picture.
But at the same time it has highlighted the points at which we must see the
events of the time as crucial in the development of Muslim religious awareness.

The reign of Mu'awiyah and, the second fitnah

Mu'awiyah (661-680) restored unity to the Arab ruling community. In fun-
damentals, he restored the system that 'Umar had created and that under
'Uthman had been adapted into a continuing political tradition, though
Mu'awiyah (Umayyad though he was) depended less on the Umayyad family
as the central support of his policy. But he restored unity no longer on the
basis of the prestige of Muhammad's city and the consensus of Muhammad's
old associates there. It was rather on the basis of a more generalized aware-
ness of common interests — together with the military force of Mu'awiyah's

2l8 THE ISLAMIC INFUSION

own loyal Syrian troops. The Arabs, aware of their precarious position in the
conquered provinces and touched by a widespread horror of dissidence within
Islam, were mostly happy to accept an arrangement which settled their mutual
rivalries on a basis of reasonably generous agreements, even though one
party, that of the Syrians, was somewhat advantaged. Mu'awiyah did not de-
pend, therefore, as had 'Umar and 'Uthman, on the inherent sanctity of the
fact that he also, like them, had had close relations with Muhammad (if only
at the last minute) ; his Syrian Arab troops honoured him personally, and were
ready to use force even against other Muslims. It was this strength that helped
persuade most Muslims to judge him the man most likely to be able to enforce
unity among Muslims and, accordingly, to give him their allegiance as Islamic
leader.

The caliphal state stood now as a more mundane imperial power, no longer
based directly on Islam. Rather it was supported internally as well as exter-
nally by a particular complex of military and physical power which was par-
tially supported in turn by Islamic faith. Militarily, Mu'awiyah could depend,
in a crisis, on the Syrian Arabs — including Christians as well as Muslims among
them — whose relatively strong discipline allowed them pre-eminence (if not
yet dominance) among all the other Arabs, The other Arab troops, however,
still formed the major part of the force of the state, and were at his disposal so
long as he controlled Syria. Fiscally, Mu'awiyah could likewise depend in a
crisis only on the revenues of Syria; but here, too, other revenues were greater.
Mu'awiyah took measures to make the central control of revenue more effec-
tive than 'Umar had made it. Already under 'Umar, the direct tax interven-
tion necessary in the Sawad of the Iraq had been leading to similar tax pat-
terns in Syria and elsewhere. There the original treaties were adjusted to a
standard level — and the poll tax of city people was graduated rather than left
as a lump sum on a per capita basis. The state was becoming more centralized.

During the fitnah wars, many Muslims had reserved the right to refuse their
allegiance to any given claimant, insisting that as Arabs and as Muslims they
could not be governed without their personal consent. When Hujr b. 'Adi in
Kufah, an ardent partisan of 'Alt, made this refusal now and on the basis of it
insulted Mu'awiyah's governor, threatening to rouse rebellion, Mu'awiyah cap-
tured him, haled him to Syria and, on his continued refusal, had him killed.
Rightly, the Kufans saw in this act an infringement of the free dignity of a
tribesman and perhaps of the direct responsibility of a Muslim individually to
God. Mu'awiyah saw it, also quite correctly, as a step essential to maintaining
intact the Islamic community.

Nevertheless, Mu'awiyah respected the freedom and dignity of Muslims
once they admitted his rule. And Islam was a cornerstone of his policy. He had
to curb equally those who would disrupt the community through asserting
local authority and those who would insist on central authority but without
the religious purpose that such authority was founded on. Mu'awiyah repre-
sented, in fact, the Muslim community as a whole, the jama' ah. As had been

THE EARLY MUSLIM STATE 210.

true from Muhammad's time on, Muslim community policy as formed by the
ruler, though not directed by the Qur'an, was still given effective support by
it. Mu'awiyah was no autocrat, but remained more the Arab chief, first among
equals. Though, like 'Uthman, he was of the Banii Umayyah (Umayyads), he
gave them no special precedence. Though he depended on his Syrians, they
received only a minimum of special privilege. His first appeal — which was in
fact what had given him victory in the fitnah — was to unity in Islam.

Mu'awiyah's reign saw the reconquest of most of Khurasan, and its per-
manent settlement with garrison towns; many further eastern Iranian lands
were systematically subdued, including part of the middle Oxus valley. Much
of Anatolia was garrisoned for brief periods, naval supremacy was maintained
in the eastern Mediterranean, and Constantinople besieged; but little per-
manent advance was achieved in Byzantine territories except for the sub-
jection of the Armenian highlands. Westward from Egypt, there was a first
occupation of the eastern Berber lands in the Maghrib as far as the present
Algeria. As was appropriate to the restorer of the caliphal state, the con-
quering force of Islam was renewed; but the expansive power was no longer so
overwhelming. A new political balance had had time to develop in such lands
as had not been overwhelmed at first, and henceforth further conquests were
due as much to the resources of an established major empire still in full vigour,
as to the pressure of the mass enthusiasms which seemed almost to carry them-
selves forward in the time of 'Umar and 'Uthman.

In his lifetime, Mu'awiyah insisted that the Muslims recognize his son
Yazid as his successor. (Yazid was probably the only man whom the Syrians
were ready to accept, for a man of any other family would bring in his own
family and clan ties, which would upset the delicate balance of forces Mu'awi-
yah had fostered.) Yazid continued Mu'awiyah's policies for four turbulent
years (680-683) hut was less fortunate in his governance. In northeastern Iran
his lieutenants continued to advance; but against the Byzantines he was on
the defensive. Just before Mu'awiyah's death, a four-year siege of Constan-
tinople had had to be abandoned with great loss. Even within Syria, some
Christian mountaineers were carrying out raids with Byzantine" support.
Yazid had to begin his reign by fortifying the Byzantine frontier. And his
appointee in the Maghrib provoked a successful revolt of the Berbers.

Yazid was soon preoccupied with the beginnings of a second round of fitnah
wars at home. The old Muslim families of Medina refused to recognize him and
encouraged resistance to him. 'All's second son and (through his mother,
Fatimah) Muhammad's grandson, Husayn, was invited to raise a rebellion in
Kufah; but then the Kufans were cowed by the Syrian governor before he
arrived. Husayn and his tiny force refused to surrender; they were isolated in
the desert at nearby Karbala.' and killed (680). Then the Hijaz itself rose in
revolt; the most prominent figure in this movement was 'Abd-Allah Ibn-al-
Zubayr, son of one of the major associates of Muhammad who had opposed
'All after the death of 'Uthman.

u<
d

43

H

ft-d

S d
s 43

t> ni

d M

<! d

' . -a

4D CD

« x!
£-23

3 .

d 43 (=1
&0 ct3 rt ni

(A h-t

<<-(

3<.a

ni >-<
u

N ,-T

CO ffl

15 g

)h d

a; r<

ft*

o C/>

•S »
•3 ,3
WE-
TS o

.2 IB

« bo

£§

<3 l-i

•° a
1

a) 73

— ctj
00 -H

co d

<+H 43

O cti

0) <3

* +>

■H rf

" "3

en co

in >>

d >,

u c(3

<d d

■' §
ffi-

43
4<
3 ,

S "d
00 43

CO l-H

o

T3

en

+J
3
ft

■■a

ST

43
3
N

* e

•8

-3 p

d

en
t>>

■s

la 2

3 3 r*

S3
CO

CM

o\

^D
I

O
00

43

a
+■>

G
O
o
a)

co

CO

43

c
>

TJ

ffi

00 i

SO ,

a

(H

3 <u

>1

^43

rti

)-c

d -m

3 HH

'42 O

42

QJ

3
N

T3

-a

3

<1) d

43 13-

o
o

43 S

w w

■S

O 43

s

"5

fa <S

*• cr
3 ^

60

'w

■^ "3

ni be
O o
o o

43
4)

en

<ai -

O

1) (U

£>

S^

II

3 s

a)

0=1

3

Si

«i

ho »

43 3

on

:s»

C/3 ho

S^;

43
o

3

a

Ph

•c

<nj
43

S d

Co -p;
I >°
8> n)
00 H
CO <1

■^ H

T3

01

s

4J

Tl

rt

0)

la

<t?

MH

0)

d

13

»

1H

ni
43

3
N

1)

3
3,

i — i

tn

ni

3

3
43
l-H

d

(fi

0J

en

t>.

CO 43

43

ft

43

d

1

o

3

d

00

CP

at **

•d "^ -S

>. ni C/)

en rt

W

.-. ni r 1 cu

CS

§

en

3
O

43

CO

CD

43

1

43

1j

-*->

o

£

cu

43

>,

>i

s

ti

d d
ft ft

1

cu

d

■3
"d

*n

S

en

d

a

■

T3

43

a

<

>

«^

»

Ss

THE EARLY MUSLIM STATE 221

This revolt had been almost crushed when Yazid died, leaving no suitable
successor in his family. His death delivered the support of most Muslims to the
most eminent of the candidates for the caliphate, Ibn-al-Zubayr, who now
appointed governors to the provinces from his capital at Mecca. But tribal
rivalries in some provinces and religious partisanship in others undermined his
authority. The local powers that acknowledged Ibn-al-Zubayr proceeded effec-
tively on their own, each chiefly for itself. In Syria, the Arab tribes were
divided into the Qays, relatively new immigrants, and the Kalb, relatively
long established in the land, and with whom Mu'awiyah had cultivated close
ties. The Kalb upheld against Ibn-al-Zubayr an Umayyad cousin of Mu'awi-
yah's, Marwan (who had been 'Uthman's chief adviser) and, defeating the
Qays, succeeded in imposing his rule in that province as a counter-caliph. 7

Elsewhere, the party of protest which had been associated with 'Ali became
active; but it too was split into factions. Though for a time some Kharijis were
willing to support Ibn-al-Zubayr, soon, in Iran and in Arabia, bands of Khari-
jis set up two separate regimes; that in Arabia controlled for a time the greater
part of the peninsula. Egalitarianism and puritanism were carried so far among
them that the ruler of the moment was subject to deposition at any time for
any moral error that the group chose to condemn. In principle, the Kharijis
made no distinction between Arab and non-Arab: what mattered was that a
man be Muslim; but those in Iran (called Azraqis) condemned as apostates all
those professing Muslims that did not accept their position; and literally put
them under the death sentence.

From the viewpoint of the non-Muslim population at large, the Khariji

7 In this case, as in some others, the historical categories used in this work differ from
the usual. Marwan is usually regarded as the legitimate caliph and Ibn-al-Zubayr as an
'anti-caliph' because in the end the Marwanids won. At the time, however, there was no
question of legitimacy, and Ibn-al-Zubayr was in fact the nearest to an effective successor
to YazJd's power, or at least to his status. Ignoring this fact has caused some authors to
misevaluate the meaning of 'Abd-al-Malik's victory, which can appear merely as suppres-
sion of rebellion. The error results from projecting backward, without warrant, an alien
notion of dynastic legitimacy.

This has been done not only in the case of Marwan, but throughout the early period:
writers have marked off the periods of the caliphate according to extraneous criteria.
Following later Sunnt Muslims, they make the reign of 'Alt (with that of Hasan) a fourth
Medina (or 'Orthodox') caliphate, set off from the reign of Mu'awiyah, who (with his son)
is lumped — as an Umayyad — with the Marwanids, though 'Uthman is not (despite his
pro-Umayyad nepotism) . For the older Muslim historians, the distinction between 'Ortho-
dox' and 'Umayyad' caliphs had a symbolic value. When 'Alt came to be lumped with the
three Medina caliphs (quite late), Mu'awiyah was correspondingly lumped with the
Marwanids. This allowed the Muslims to split the work of establishing the caliphal struc-
ture into two parts: into the 'good' side of that work (including whatever was approved,
of the work of subsequent caliphs), which was ascribed to 'Umar; and into the 'bad' side
(including much of what 'Umar did), symbolized in the setting up of 'kingship', which
was ascribed to Mu'awiyah and the 'Umayyads'. For this purpose, 'Uthman was 'Ortho-
dox' and not 'Umayyad'. But such considerations need not bind the modern historian.
They are of the same order as the inclination to see as 'heretical' any forms of Islam which
were not later received by the majority (or rather by certain widely respected later
Muslim authors).

222 THE ISLAMIC INFUSION

form of Islam may have seemed the ideal form of Muslim rule: the Kharijis
remained separate from the dhimmi populace and in maintaining the purity
of Muslim life ensured clear control over the Muslims in the lands, who did
not become rivals to the local establishment. They were lined up against the
garrison towns and, at least in the following decade or so, received a degree of
support (at least in the Jazirah) from the rural population. At any rate, they
were able to carry on in the Jazirah and in Iran what amounted to guerilla
warfare. During the second fitnah, it was the Kharijis who succeeded in con-
trolling the largest extent of territory, though they did not control any of the
important garrison towns. But a more or less passive support by the peasantry
could not make up for lack of support among the organized Muslims (except
among the Arabian tribes themselves, who, however, were not inclined to
allow the Kharijis an overruling power, to escape which some of them may
have supported the Kharijis against the city Muslims). The troops of Basrah,
with some support from elsewhere, made steady progress against them.

In Kufah, a quite different party of protest prevailed: that of 'Alid loyalism.
It proved less conscientious than the Kharijis and also less intolerant. Repen-
tance for the failure to support Husayn at Karbala.', whose death was all the
more horrible because of his descent from the Prophet, moved many to at-
tempt some form of expiation. Those who were actively loyal to the cause of
'All's family came to be called the Shi'ah (party) of 'All. A number set off to
avenge Husayn's death against the Syrians, without positive results. This
sentiment soon was channelled into an attempt to set up another of 'Alt's sons,
Ibn-al-Hanafiyyah, as caliph. This Shi'i revolt was led by Mukhtar b. Abi-
'Ubayd, who attempted an egalitarianism more moderate than that of the
Kharijis. 8 He gave the non-Arab converts to Islam, called Mawali, equal status
as to booty. But this so enraged the old Kufan families that they turned
against him. Even so, he was put down (687) only with difficulty by Ibn-al-
Zubayr's governor at Basrah, who diverted Basran energies for the moment
away from the Khariji campaigns, to this end.

Each of the major claimants to the caliphate hoped to control the whole of
the Muslim territory, no one province being conceived as capable of standing
alone. Of all the conflicting forces, the Umayyad house in Syria proved the
strongest. As we have seen, those who were keeping the Kharijis at bay in the
Iraq accepted the Zubayrid leadership and under it overcame the Sht'is in the
Iraq; but Ibn-al-Zubayr, head of his party, remained isolated in the Hijaz,
partly because of Khariji control of so much territory in Arabia. Meanwhile,
those who counted in Syria proved more able to unite than did the Iraqis;
Egypt had quickly fallen to Marwan and his son, who thus succeeded to the

8 The form Shi'i as derivative of Shi'ah is awkward to pronounce in English and some-
times even those who usually use the -; ending, instead of the older -He, will say 'Shi'ite'.
In India, a Shi'i is commonly miscalled 'a Shi'ah', but properly a Shi'ah is a. party, not an
individual; if one says 'the Shi'ahs did' such and such, this should mean that the several
Shi'i parties did so — the 'Twelver' Shi'ah, the 'Zaydi' Shi'ah, etc. — not the Shi'is as
individuals.

THE EARLY MUSLIM STATE 223

home provinces of the Syrian caliphate. In the subsequent contest between
Syria and the Iraq, Syria won. The forces of the garrison towns then continued
their struggle against the Kharijis as readily under vigorous Marwani leader-
ship as under the loose authority of Ibn-al-Zubayr. (I use the general term
'Marwani' for the men associated with Marwan and with his descendants, the
'Bami Marwan' or Marwanids.) Thus eventually the Marwani forces disposed
of all their rivals; they took Mecca in 692 and put an end to Ibn-al-Zubayr
himself. {In the process, the Ka'bah was wrecked and had to be rebuilt.)

The Marwani state

It was Marwan's son 'Abd-al-Malik {692-705) who thus became the third great
caliph in Islam, after 'Umar and Mu'awiyah. The fitnah had not been brought
to a conclusion by an accommodation, this time; it had been fought out to the
end. Accordingly, 'Abd-al-Malik had to establish the state unambiguously on
the basis first of all of force, with religious allegiance brought into play only
when force had first decided who was to be master. Supreme personal power
was made hereditary in Marwan's family {succession by designation of the
previous ruler, in effect), as can be seen in the chart of the Umayyad family.
'Abd-al-Malik's lieutenant, al-Hajjaj b. Yusuf (d. 714), who had reduced
Mecca to obedience (it was he who had bombarded the Ka'bah), ruled single-
handed the eastern half of the empire — what had been the Sasanian dominions.
A former school-teacher of the Thaqif tribe (of Ta'if), he had risen by his ruth-
less efficiency. He added to the revenues by administrative vigour which
regularized and increased investment in the agriculture of the Sawad of the
Iraq, and he controlled the Iraqi Muslims by frank terror directed against the
disaffected. After further Iraqi revolt, he built a new provincial capital, al-
Wasit, between Kufah and Basrah, handy to either centre of opposition, and
garrisoned it with loyal Syrian Muslims; these were kept jealously separate
from the Iraqi Muslims. In the western (ex-Roman and Arabian) territories
'Abd-al-Malik (and his brother, 'Abd-al-'Aziz, in Egypt) ruled with policies
equally firm, though terror was less needed. (In such territorial arrangements
we may see the last traces of the Meccan system as Muhammad had recon-
stituted it. The lands from Syria to the Yemen were still the core of the em-
pire, though now ruled one-sidedly from Syria; while the former Sasanian ter-
ritories were ruled as a huge appendage thereto.) 'Abd-al-Malik was succeeded
without question by his son al-Walid in 705, and then by other members of his
family (the Marwanids), for almost half a century.

The Marwani state was, however, thoroughly Islamic as Islam was then
understood. In the first generations, Islamic religion was, by later standards,
still rudimentary in the consciousness of the faithful: it was above all a badge
of united Arabism, the code and discipline of a conquering elite. On this basis,
'Abd-al-Malik and his family upheld the ideal of jama'ah, of the solidarity of
the Muslim community over against the factionalism of the Arab tribal or

<<

8

rt |

•a o

eg B
co^.

.- >>

<!>

A

M

a

ii

ni

>

o

01

en

-P C n!

<3
3

in
.&
U

11

H

■s

IS

<:

- a

o

Pi

CO

•n

ID
I

*o

ici

5

- 3-
i

<

a

n

a

«!

El

a

o

s

«

-<!

n.

»

w

£3

w

S

H

•o

■a

'I

6
P

is

CO

oo

A

00

- h -

I

■>*■

u

Z
<<

s
x

H

43

P4

<3

-«

i

a

CO
00

W T3

- N •«
cS O

* >

o

O

ft

>,

M

W>

< w

r

o

T3

>

43

o

<! o

o

N

H

u

< u
X. »

Q a

« »

< be

I

N

h
-S-

«!
B

M

Ti-
ft

«5og

— «
■

"3

i-l

<

O-l

1-2

i4

i-i

t-

Q

t*

►4

►-(

5

(S

$

><

a *
d

cr 1 d

ID O

to 5

& 1

IP

« o
*"§
II

3^

Pi fl

o
ft

o

Hi

M

is

c
2 8

^ g is

ft rt <d
"> -3-2

o.ll

■Vot

-3 a g

J3 pi
D-i ® erf

f "S i

w

H
O

a§

■s

a
-a

226 THE ISLAMIC INFUSION

regional groupings. 'Abd-al-Malik and his lieutenant al-Haj jaj were concerned
to maintain the supremacy of Islam when they substituted for the coinage of
the old infidel empires new coins with Islamic inscriptions, and intervened to
assure fidelity in the reciting of the Qur'an by encouraging more exact ways
of writing it down than an imperfect script had hitherto allowed. Alongside
the governors, they appointed special judges in the camp towns, called qd^is,
who were to settle disputes among the Muslims on an Islamic basis. They did
not encourage, and even discouraged, conversions to Islam from the subject
populations; but this was in conformity with the most common view of Islam
among Muslims. Islam, among the several revealed religious allegiances, was
the one that should guide those in command among men, and these should be
the Arabs, to whom Islam was properly given.

The Arab realm made its last major advance under 'Abd-al-Malik and al-
Walid (705-715). The losses suffered during the second fitnah were recovered,
notably in the Oxus valley and among the eastern Berbers. The Berbers, the
bulk of the inland population in the Maghrib (North Africa), were rather like
the Arabs themselves in having been marginal to citied civilization, though
their mountain nomadism was less unsettled; they had little confessional reli-
gion and were converted en masse to Islam. Once they had had to admit the
superior power of the Arabs, they joined them in the further conquests. The
more westerly Berbers were relatively easily swept up into the movement.
Thus a secondary centre of Muslim conquest was established in the west
Mediterranean, in some respects relatively independent of the main centre.
By 711 a raid was launched into the Spanish peninsula, with the aid of newly
converted Berber troops, and that kingdom (like some other Christian lands
subject to a fiercely persecuting church) was quickly and definitively
conquered with local Maghribi resources and little help from the east
Mediterranean.

Expeditions from Syria against the heart of the Byzantine empire led again
to a great siege of Constantinople, but when that failed no part of Anatolia
beyond the Taurus mountains could be held. (The Byzantine cities in the
Maghrib, however, were occupied when the Berber tribes were converted in
their hinterland.) Eastward, the lower Indus valley, Sind, was conquered
partly by land and partly by sea. The local Buddhists, who were evidently in
part mercantile in orientation, seem to have preferred the Muslims to the
Hindu ruling classes.

More important was the occupation, from Khurasan, of the Oxus and Zaraf-
shan basin to the northeast under the able lead of the general Qutaybah b.
Muslim. There, mercantile city-states tried to play off, for a time, Chinese
against Muslim influence, but in the end were forced to accept Muslim control.
The frontier between Muslims and Chinese was finally settled in the high
mountains just halfway between the capitals of the two imperial powers. 9 The

* The best introduction to the nature of Muslim conquest, after the very first thrusts,
is H. A. R. Gibb, The Arab Conquests in Central Asia (London, 1923); it brings out the

THE EARLY MUSLIM STATE 227

Muslim conquests, like any conquests, had their shameful episodes. The story
goes that when Baykand (near Bukhara) was occupied by Qutaybah's men,
the captain in charge took possession of the two beautiful daughters of one of
the citizens, and ignored the man's complaint that he should be so singled out;
whereupon the man stabbed the captain. In any case, the rest of the region
being as yet unoccupied, when the main part of the Arab forces had withdrawn
southwards, the town rose against the Arab garrison and drove it out, pre-
sumably regarding the Arab incursion beyond the old Sasanian territories as a
transitory event. But the town was immediately reoccupied and made an
example. The captured men were all killed and the women and children en-
slaved, and the city itself was levelled. (Yet this ferocity was mitigated — and
the mitigation is doubtless as indicative as the ferocity. For, as it happened,
a large part of the merchants of the city — -the main body of its inhabitants —
had been away on the trade caravan eastward; when they returned in due
course, they were allowed to ransom their wives and children, and eventually
they rebuilt their town.)

Under the later Marwanids some of these acquisitions were further en-
trenched (notably in the face of a rebellion in the Zarafshan valley [a northern
near-tributary of the Oxus] which threatened Muslim holdings in the whole
region, crushed in 737). Some lesser bases were acquired, for instance in
southern Gaul. But on the whole, conquest came to a stop. The whole of the
Sasanian and half the Roman empires had been conquered, together with
several lands which had been in the cultural orbit of one or the other empire
but had not recently been under their sway.

The Arab tribesmen had established themselves in the occupied territories
as rulers and as ultimate dispensers of the revenues of the land (that is, accord-
ing to ways dating from the time when cities were first established, whatever
produce could be taken from the actual producers without incapacitating
them). They had left the internal life of the conquered Christian, Jewish, Maz-
dean, and Buddhist communities to proceed on its own, provided the supre-
macy was left to the Arabs. For the time being the prevailing culture con-
tinued to be Hellenistic, Sasanian, or whatever existed locally, while the Mus-
lim Arabs themselves carried with them as much of the old Arabian culture as
could survive transplanting.

This Arabian culture possessed, in fact, considerable vigour. It was carried
and reinforced by a reconstituted tribal system, based on the garrison towns
(and not very influential among those who had remained Bedouin in Arabia).
On the imperial scale, the smaller tribal units had lost their importance and
tribal groups tended to coalesce into larger alliances. In each garrison town or
occupied country, two or three main tribal blocs were formed, which in turn
recognized allies in corresponding blocs elsewhere. In a general way, they

complications introduced by a local setting and by long-distance connections, and the
special role played by Islam as such. It also shows how the evidence can be used to cut
through legend to actuality. Though it is limited to one area, it is suggestive for all.

fc

H

S

W

J* 1

O

<g

-

-»-»•

f,

.J-1-

K

d

H

a

K
H
D
O

l-

a-

c! 2 u
« p A IS

S.s

C4 rt ffl

•a -="

3

a-

en

«
H
K
O

n
<!

13

£

a

■a-.

T3

-<c3

ni

- s
a

X!

S3 <u

'Am
intl
Iraq

1

'Abd al-
in N.E.
Arabia

Taghlib
in the
Tazirah

ffl.S £

u

-4-J

XI

nJ O
s-< a

«.g»

~ 0* s-i

H oH-

H -S m

-3

™ ^ ai

O* oS

THE EARLY MUSLIM STATE 229

identified themselves (from the time of the struggle between the Qays and the
Kalb in Syria) as the 'northern' (Nizar) Arabs, represented by the Qays,
against the 'southern' (Yemeni or Qahtan) Arabs, identified with the Kalb.
(This was an ancient division, having long ceased to have much geographical
significance, which was given a largely artificial genealogical form.) At least as
important was a division within the 'northern' bloc between 'Mudar' tribes
and 'Rabi'ah' tribes. Tribes affiliated together as Rabi'ah and as Qahtan ten-
ded to be those which had long been associated with the settled populations in
Syria and the Iraq; many of them had been Christian. Rabi'ah and Qahtan
found themselves allied together more often than did Rabi'ah and Mudar.
While some important non-Arab groups were associated with Mudar tribes,
especially with Tamim, it was eventually with movements more or less associ-
ated with Qahtan and Rabi'ah that the ordinary non-Arab Muslims threw in
their lot.

It was in terms of this neo-tribalism that power struggles within the Muslim
group took place: they were carried on in factional fighting among the Arab
tribal blocs. It was also on the basis of this neo-tribalism that elements of the
old Arabian culture were carried over into the conquered lands. Without the
all-Arabian, Bedouin-based common patterns, even the force of Islam could
probably not have forestalled a renewed Aramaism or Hellenism such as had
absorbed earlier Arab conquerors in the Fertile Crescent.

These Arabian traditions had relatively little inherent connection with Islam
itself. Islam possessed, indeed, enormous dynamic force as a cultural tradition.
But it could offer little, at first, toward the settlement of day-to-day social
questions. Here a common body of cultural patterns which the conquerors
shared as Arabs served better. In the new garrison towns, being an Arab was
as important, therefore, as being a Muslim. Christian Arab tribes which had
joined in the conquest were regarded first of all as Arabs, and not classified
with the conquered dhimmi Christians. It was in no way anomalous that one
of the greatest poets of the court at Damascus was al-Akhtal (c. 640-710), a
Christian Arab, panegyrist of 'Abd-al-Malik. The Marwanids, as had Mu'awi-
yah, continued to see their power as based on agreement among the Arab
tribes, rather than on individuals. Hence to be a Muslim in the full political
sense, a convert had to become associated, as client {mawla, pi. mawdli), with
one or other of the Muslim Arab tribes; as such, he and his descendants were
socially inferior to the original members of the tribe, but shared its allegi-
ances. Hence even the new converts expressed their interests through the
Arab factional fighting. In this way, a base was maintained for the expression
of the old Bedouin ideals, passed on even to the new recruits to Islam.

When the society of the garrison towns was well established, the classical
Arabic poetry of Muhammad's time was vigorously transplanted to the new
setting; there, as in the Bedouin society, it was in the service of tribal pride or
ambition. (Bedouin poetry continued to be composed within the Bedouin set-
ting itself, of course, but it rested no longer on financing from the old rival

230 THE ISLAMIC INFUSION

empires, and henceforth its cultural importance there was less dynamic.)
Much poetry served to express memorably the new, large-scale tribal loyalties
of the garrison towns. Poetry was adapted to a newer type of loyalty also,
religious loyalties; some of the best poets were zealous Kharijis and Shi'is. The
two greatest poets after al-Akhtal were al-Farazdaq (d. 728), a Shi'i, and Jarir
(d. 728) ; their rivalry was celebrated throughout Islamdom and involved all
the Arabs in partisanships that crossed tribal lines (for both were of the same
tribe). As before Muhammad, the poetry was expressed in the standard
Bedouin tongue (to which the Qur'an also approximated). This became the
regular language in which Muslim religious reports, legal decisions, and the
like were recorded, to the exclusion not only of any particular tribal dialects
(even Quraysh) but also of the uninflected dialect associated with the 'settled
Arabs', which rapidly came to be more commonly spoken in the camp towns. 10
Thus the best in the pagan Arabian tradition was preserved, with its language,
and adapted to the new urban and Islamic conditions. It was thus assured a
share in the new culture to be created.

Under the Marwanids, what had been a collection of occupied territories
was gradually transformed into a relatively unified empire, the whole appara-
tus of the rule of which was taken over by the Arabs and run in Arabic. At the
same time, conversions to Islam from among Aramaeans and Persians were
becoming ever commoner, even while much of the administering personnel
continued to be non-Muslim. On the other hand, the Arabs were learning how
to live like eastern Mediterranean or Iranian landlords to some degree.
Distinctions inherited from older political conditions, such as the autonomy
of the formerly Sasanian areas as a body, tended to disappear; governors
were appointed to every province directly from the capital. The Arabs ceased
to be occupying troops and became the ruling stratum among the rest of the
population, which was gradually being assimilated to them as they in turn
adapted themselves to its expectations. A new civilization, common to all in
the region, then began to take form.

10 J. Blau, 'The Importance of Middle Arabic Dialects for the History of Arabic',
Studies in Islamic History and Civilization, ed. Uriel Heyd (Hebrew University, 1961),
pp. 206-28, points out how early sedentary Arabic was, as a non-inflected language.
It has been speculated that even the pre-Islamic Quraysh used a non-inflected language
such as some Syrian Arabs may well have used, but this seems highly improbable. The
psychological gap between an inflected and a non-inflected language is greater than some
students of the problem seem to suppose, masking the reality for themselves by using the
word 'colloquial' where they mean 'vernacular'. The point of departure in these studies is
Johann Fuck, Arabiya: TJntersuchungen zur arabischen Sprach- und Stilgeschichte (Berlin,
1950). which collects information on the fate of the classical Arabic from the time of the
conquests to that of the Seljuqs. Cf. also Chaim Rabin, Ancient West Arabian (London,
195 1), for dialect study.

BOOK TWO

The Classical Civilization of the

High Caliphate

Thereafter [the Tin Woodman] walked very
carefully, with his eyes on the road, and when
he saw a tiny ant toiling by he would step over
it, so as not to harm it. . . . 'You people with
hearts,' he said, 'have something to guide you,
and need never do wrong; but I have no heart,
and so I must be very careful . . .'.

— L. Frank Baum

PROLOGUE TO BOOK TWO

With the caliphal state established as an enduring political structure within
whose framework the high-cultural life of the region was to be carried on, we
come to the time when the Islamic impulse began to have at least a condi-
tioning and limiting, at best a positively formative effect on all aspects of the
Irano-Semitic cultural life. We enter the time of Islamicate civilization.

The refonting of traditions

From about 692, the caliphal state was a well-established agrarian-based em-
pire and it continued so till about 945, when the caliphal government became
subordinated to other powers. We may call this the 'High Caliphate' in distinc-
tion from the more primitive caliphate from Abu-Bakr to Mu'awiyah, when
the character of the state and its durability were still in question; and in dis-
tinction from the latter-day 'Abbasi caliphal state after 945, which was some-
times only a form carrying a figurehead, and was at best a local power with
special prestige — as well as in distinction from other, less universally recog-
nized governments claiming caliphal status. (This implies a periodization at
variance with that which has been conventionally used. The chart on
comparative periodization will show how the two systems mesh.)

The period was one of great prosperity. It is not clear how far this was the
case throughout the Afro-Eurasian Oikoumene, but at least in China at that
time what may be called a 'commercial revolution' was taking place. Under
the strong government of the T'ang dynasty, which had reunited the whole
Chinese agrarian region and maintained internal peace there and power abroad
till near the end of our period, commerce became much more extensive and
more highly organized. The intensity of commerce within China and of the
accompanying urbanization greatly exceeded the preceding peak, reached in
the immediate post-Axial centuries, when it was on a level with that of the
contemporary Hellenistic Mediterranean. The Chinese economic activity was
directly reflected in the trade in the Southern Seas (the Indian Ocean and
seas eastward), where Chinese ports became an important terminus for Mus-
lim vessels. More generally, at this time began a long period of limited but
unmistakable Chinese cultural ascendancy in the Oikoumene as a whole — re-
placing the recent ascendancy of Indie impulses and the still earlier ascen-
dancy of Hellenism. In art and technology, new ideas came with more persis-
tence from China than from any other one source; and the image of a well-
governed Chinese empire became well-rooted abroad, at least in the Muslim
literature.

It can be surmised that the commercial life of the lands of Muslim rule was
given a positive impetus by the great activity in China, especially considering
its important connections with China both via the Southern Seas and overland

233

a

•d

-0

o

rt

«

+3

E

N

6

13
o

A-

XI-

n

s

«

S

s

<u

4->

Pn

ni

Hi

. ,

o

o

A

,R

a
o

■a

o

O

.ft

15

.ft.

13

a

u
ft

ft

o

o

>

R

rt

13

>-,

a

ctl

R

O

>.

o
O

o
o

0)

T3

5

s

s

O

&

tn
tn

CD

R
u

ft

r 43

a

■a

o

fi

■a

.o

13
CS

6

J-
I

ffl

o
m

-0

0)

I

.3

■a a

MP

-R
<d ,g>

?!

ID w

in vo

ft

-R
ft

o

Ph

R
j3

P

o

i

P

,R
ft

S

1

■Si-

,R
ft

"3
o

:§

<1

-a

,R
ft

13
<u

'ft

3

13
T3

R
o

"ft

•R
ft

D N

O

o o

d
O

Ah

-R.2

■a

d
ft

<d

JO
<

d

►-5

13
O
'u

111

O

be
R
O

R

d
+■>

T)
d
T3

■a

d
M

CO

PROLOGUE TO BOOK TWO 235

through central Eurasia. In any case, commerce also enjoyed the great bene-
fits of an extended peace which the caliphate was able to ensure within its
own domains. In contrast to the disastrous warfare between Byzantium and
the Sasanians which had destroyed cities and ruined agriculture again and
again here and there in the Fertile Crescent, that central area was seriously
ravaged only toward the end of our period. Even when the west Mediterranean
lands of Islamdom became independent, as they soon did, fighting between
them and other Muslims was minimal. Muslim navies dominated the whole
Mediterranean. The caliphate limited its military endeavours to annual ex-
peditions against the Byzantine empire and minor operations along the other
northern frontiers, where its power was enough concentrated, effectively to
overawe any threats. Internally, disturbances were relatively infrequent and
generally localized. The scourge of warfare was kept in check in most places
most of the time during almost the whole of the High Caliphal Period.

Under the Marwanid caliphs and especially under the 'Abbasids who suc-
ceeded them, the barriers gradually fell away that had kept the evolution of
the cultural life of the several conquered nations separated from each other
and from the internal development of the Muslim ruling class. The leading
social strata of the empire, of whatever background— even that minority that
was not yet becoming Muslim — lived in a single vast society. Their common
cultural patterns formed what can be called High Caliphal civilization.

These cultural patterns continued to be expressed, till almost the end of
the period, in terms of a variety of linguistic and religious backgrounds. Syriac
and Pahlavi continued to be major vehicles of high culture along with the
newer Arabic. The revivifying of the Hellenic intellectual tradition, a most
striking feature of the period, was marked by Greek translations into Syriac
as well as into Arabic; Christians, Mazdeans, Jews, and even a group of Hel-
lenistic pagans (at Harran in the Jazirah) alongside Muslims, shared in many
of the concerns of the time either within their own religious traditions or
across religious lines, often in co-operation with each other.

Nevertheless, it was under the common administration and protection of
the Muslim caliphate that the society prospered and the civilization flowered.
What brought all the traditions together increasingly was the presence of
Islam. Mazdean polemics, the ecclesiastical organization of the several Chris-
tian communities, Jewish commercial activity, all reflected the effects of the
caliphal power and sometimes even the challenge of the Islamic sense of the
divine. Not only was the common government, which all communities had
learned to respect, Muslim. 'Mudari' Arabic, which the early Muslim com-
munity had used, was the one generally recognized language; it rapidly be-
came the common language of administration and (in a form more or less
touched with sedentary dialects) of longer-range commerce; accordingly, all
men of wide ambition had to learn it. Those who became members of the
Arabic-using, and still more of the Islam-professing, group found themselves
in an advantaged position, not only politically and socially but intellectually.

236 THE CIVILIZATION OF THE HIGH CALIPHATE

This position became steadily stronger as the various cultural traditions, even
in many cases religious traditions and still more so those traditions less direct-
ly associated with religion, were rendered into Arabic by translation of texts
and by other transmission of skills and of lore. Eventually, he who had access
to what was in Arabic had access to all the regional heritages. And once the
traditions were carried in Arabic and their carriers became Muslims, the Is-
lamic impulse began to penetrate them and to inform them with its spirit.
By the end of the period, the many older Semitic and Iranian communities
had given way to a single comprehensive Muslim community. The representa-
tives of a strictly Mazdean or Syriac Christian culture had become few and
those who remained were ceasing to be culturally productive. The living
traditions had been largely brought together and transformed under the
auspices of Islam. What had been lost in the process^-especially on the level
of religious doctrine or of literary forms and ideals ; — was supplied from the
older Arabian heritage as it had been reconstituted in the early Muslim com-
munity, and most especially from the norms and expectations which had been
developed by Islam itself both in the time of Muhammad and in the following
centuries as it took up and continued the old monotheistic traditions of the
region. The significant cultural life of the region was flowing within the frame-
work of the Arabic language and the Islamic faith, among Muslims and such
adherents of other religious traditions as had been assimilated culturally to
Muslims.

On continuity and cultural florescence

The process of Islamicizing the traditions had done more than integrate and
reform them. It had released tremendous creative energies. The High Caliphal
Period was one of great cultural florescence.

Before quite Modern times, in civilized societies, a sense of continuity and
repetition of established patterns moulded overwhelmingly the channels with-
in which historical life flowed. The dialogue through which traditions were
developed moved cautiously and did not readily introduce radical changes.
Wholly new traditions were few. Always people were confronting new situa-
tions and inventing new ways; initiative and originality were always praised,
if they were on a high level and in good taste. But 'good taste' was felt, usually,
to exclude overly drastic innovations.

Continuity is easily explained. There was in fact a limit, under agrarianate
conditions, to the amount of innovation that could effectively be absorbed.
There was a limit, for instance, to the amount of capital available for invest-
ment; the amount was tied closely, ultimately, to the agricultural surplus.
Any innovation that required great capital outlay was normally ruled out in
advance, and this limitation alone was very far-reaching. People could not
afford to tear everything down, retrain all the social cadres, and build anew.
But apart from such considerations, there was an inherent tendency in style

PROLOGUE TO BOOK TWO 237

which militated against radical innovation. As particular forms of cultural
expression developed richly, individuals found they could rise to more subtle
and more refined levels if they worked out more fully the possibilities of an
already well-developed tradition, than if they struck off crudely on their own.
Where patterns of cultivated awareness already existed, they were rarely
sacrificed to private experiment.

Hence great cultural florescences were fairly rare: that is, times of general
cultural creativity and innovation, when relatively independent literary,
artistic, social, and religious traditions were launched afresh over the whole
range of culture. Even periods in which major creativity was dominating some
but not other aspects of culture were less common (even in the core regions)
than periods with only relatively secondary sorts of innovation, within estab-
lished traditions. When overall florescences did occur, of course, they marked
the course of world history. The 'Axial Age' had seen such major creativity in
all the major core regions. India had evidently known another such period in
the early centuries of our era; by the time of the advent of Islam, the results
of this latter Indie creativity were being felt very widely across the hemisphere
in art, science, and religion, as well as in more trivial ways. In the Islamic
period, T'ang (618-907) and Sung (960-1279) China saw such a florescence.
Western Europe was to go through times of florescence somewhat later —
florescence which, in fact, was ultimately to lead into the Modern world-wide
cultural transformations. The High Caliphal cultural ferment too formed such
a florescence and its outcome largely dominated world history for centuries
after. To be sure, even these outstanding periods of creativity built upon ele-
ments of received knowledge and expectations, if only by way of revolt. Even
lines of the most radical innovation presupposed the primacy of continuity.

A conflict and contrast of traditions within a single social group naturally
offers opportunities for creative ventures into new forms not fully grounded in
any one tradition; certainly the existence of such contrasts was one of the
several circumstances which made possible a florescence in early Islamic times.
We cannot look on it as a full explanation of the florescence; for cultural crea-
tivity may arise with a minimum of conflict of cultures, being set in motion
rather by internal impasses when all old ways seem to break down, or by
internal openings when old limitations become irrelevant; whereas sometimes
a conflict among traditions has no creative outcome at all. Yet the High Cali-
phal culture took creative form, as it happened, very much in terms of the
conflict and mutual accommodation of cultural elements of diverse origins.
It was here in large measure that the creative opportunities lay which people
then realized.

Before Islam, the various sorts of tradition had had each its stronghold in
its own sphere — not only in this or that phase of culture, in art or govern-
ment, etc.; but also, to some extent, within the lines of this or that religious
community. With the coming of Islam, the older balance among types of
traditions was upset and new elements were juxtaposed. Islam itself had set

238 THE CIVILIZATION OF THE HIGH CALIPHATE

going new traditions, relatively independent of any older heritage, with which
people had to come to terms. With the inclusion of areas in the empire which
had not originally belonged to the Sasanians, their local traditions were
brought into the melee. Perhaps above all, all traditions had to be refonted
within the loyalties and symbols of a single community: a major change in the
life of the region. Insofar as each major tradition of thought and expectation
was potentially applicable throughout all of life, the several traditions came
into conflict once the old compromises that had delimited spheres of action
among them were broken down. Here lay the necessity and opportunity for
basic innovation. In High Caliphal times a new balance was worked out, in
which again each of the various orientations, transformed, had its own sphere
of primacy.

Rival cultural orientations under the High Caliphate

Certain fields of thought and practice came to be dominated by piety-minded
representatives of the Islamic hope for a godly personal and social order — a
hope inherited from the Jewish and Christian priests and monks and rabbis
and their flocks. Muslims learned something from these latter, but proved very
independent in giving the hope an integrally Islamic form. Every individual's
life should be directly under the guidance of God's laws, and anything in
society not clearly necessary to His service was to be frowned upon. Among
both Sunni and Shi'i Muslims, a host of pious men and women who came to be
called the 'ulatnd', the 'learned', worked out what we may call the 'Shari'ah-
minded' programme for private and public living centered on the Shari'ah
law. As might be expected, these 'ulama' scholars dominated Muslim public
worship. They exercised a wide sway, but not exclusive control, in Muslim
speculative and theological thought. They exercised an effective — but never
decisive — pressure in the realms of public order and government, and con-
trolled the theoretical development of Muslim law. The fields of Arabic gram-
mar, of some sorts of history, and even sometimes of Arabic literary criticism
were largely under their influence. But many aspects of culture escaped their
zealous supervision — including much of the Islamic piety of the time. It was
probably only among the merchant classes that their ideals came close to over-
all fulfillment. Their work, however, gave a certain dignity to the whole social
edifice. As a whole, that edifice reflected the aspirations of the 'ulama', and
the intellectual and social patterns that followed therefrom, more than it
reflected any other one set of ideals.

Two other religious orientations then found less widespread expression in
Islam. The ascetic and monastic tradition within Christianity (and probably
Buddhism) found its counterparts among Muslims, called sufis, who learned
as much from their predecessors as did the Shari'ah-minded, but had a rather
less comprehensive programme for other than contemplative aspects of life. A
second movement, that of the Batiniyyah, with more immediate political im-

PROLOGUE TO BOOK TWO 239

plications, appealed to a tradition of esoteric love which may also have had
pre-Islamic roots. It proved potent at moments of social protest.

Set over against the ideals especially of the Shari'ah-minded Muslims was
what may be summed up under the heading adab, the worldly culture of the
polite classes. While the Muslim courtier, administrator, or intelligent land-
owner paid due honour to the aspirations of the professional Muslims, most of
their efforts were devoted to living out a very different pattern from what the
latter approved. Their etiquette, their conversation and fine arts and litera-
ture, their ways of using poetry and music and even religion, and their whole
social pattern of position and privilege, with its economic and political institu-
tions and its politics, formed a distinct set of genteel standards, prevailing
among Muslims and non-Muslims of wealth and position. These standards,
this adab, spread from one end of the Islamic domains to the other; fashions
would most commonly be created in Baghdad, the most important seat of
government, and would be eagerly adopted everywhere else. In this adab cul-
ture ancient pagan Arabian ideals had their place, a poetry inspired thereby
was the highest gift, and the Arabic language was supreme; but a greater
place on the whole, even in literature, was reserved for the Iranian imperial
tradition of Sasanian times. Both elements were consciously adapted to Islam.

Independent, both of the prophetic-monotheistic and of the imperial tradi-
tions, was the highly self-conscious tradition of Falsa/ah. This was an inclusive
term for the natural and philosophical learning of the Greek masters. Some
other elements from the Greek traditions had a place in the developing Islami-
cate culture, but it was only in this intellectual sphere that Greek tradition
was supreme. Already before Islamic times, science had been the Hellenic
stronghold in the Fertile Crescent and Iran. Several sciences which depended
exclusively on natural reason for their development, including speculative
philosophy, were carried over in a body from the Greek-Syriac tradition.
Writers such as Plato, Aristotle, and Galen were the acknowledged masters.
When the main classical works had been translated into Arabic they became a
primary part of the intellectual equipment of a class of students (Muslim,
Christian, and Harrani pagan) who, while never fully accepted by the Shari'ah-
minded Muslims, were nevertheless respected by society at large. They
flourished especially in their role as practitioners of medicine or of astrology,
two of the most popular sciences. In principle, however, they too had more
comprehensive claims on life, which rivalled both the Sharfah-mindedness
of the 'ulama' and the adab culture of polite society.

A caveat on our use of periodtzation

When we are studyi' g a total civilization, we can no longer follow a single
chain of events from key point to key point, as is possible when we are tracing
the origin of particular components in a culture. We must see how the society
developed simultaneously in many parallel and interconnected spheres. We

24O THE CIVILIZATION OF THE HIGH CALIPHATE

must take whole chunks of time at once, that is, presenting first one sort of
activity, then another, in the given period. But since the culture constantly
changes, to relate the art or the religious thought, say, of any particular time
span to other events in that time span means to see that art or that thought in
a special way. We must stress what most stands out in that particular time
span. For instance, when the unit for comparing different facets of culture is
the two and a half centuries from 700 to 950, such continuing movements as
the Islamic religious partisanships of the first decades loom much larger than
the relatively transitory Arab tribal feuds of the same decades, though the
latter filled the chronicles. Accordingly, our choice of periodization will help
determine the weight we give to various events and hence the sort of picture
we get of the civilization as a whole.

The selection of time spans must be anyway somewhat arbitrary — even
more so than the setting of demarcation lines between one region and the
next. In some respects, the shorter the intervals chosen, the more precise can
be our understanding of developments. But in a detailed study by decades or
by generations one may not see the forest for the trees. The autonomous
internal development of the various traditions that make up a culture comes
clearer if we take as our periods one, two, or even perhaps three centuries to-
gether. In Book Two we will concentrate on those social pressures, those cur-
rents of thought, those lines of tradition, that were of relatively continuing
significance throughout the High Caliphal Period. Some of these, for instance
natural science, we will treat mostly as a single development for the period as
a whole. Yet even the movement of natural science was dependent on other
movements and affected them in turn. Within the longer period, we will in-
terrupt the various separate movements so far as feasible — and not only the
political ones — so as to trace the interactions that took place among them
within much shorter time spans as general cultural conditions evolved. More
than in any other period before Modern times, intellectual, religious, social,
and political developments were then closely correlated within fairly short
time limits.

i i g

The Islamic Opposition,
692-7^0

From the time of the Riddah wars, at least, the Muslim community, or its
leaders, had undertaken major political responsibilities as an essential conse-
quence of their faith. Under the conditions of Marwani rule, those who were
most serious about Islam were also most serious about the political responsi-
bilities which an acceptance of Islam entailed. Inevitably, a sensitive con-
science found that much was lacking. It was almost a corollary of the political
responsibility called for by Islam that the tradition of faith proved to be
developed most actively in an atmosphere of political opposition to the ruling
forms.

Meanwhile, the Muslim rulers were governing not only the Muslim group —
piously conscientious or not — but also the general population of the conquered
regions. These too had their ideals; and among them a most potent ideal was
the longing for a strong central authority which would suppress warfare and
defend the rights of the weak. It was such demands the caliphs found easiest
and most expedient to move toward satisfying. Gradually, the ideal of a bene-
volent absolutism attached itself to the caliph's court, confronting the ideal of
an Islamic egalitarianism in the opposition. Even in Marwani times the con-
trast became increasingly noticeable.

The caliphal state approximates an absolute monarchy

From the point of view of leading classes among the conquered peoples, the
rule of the Arabs had been acceptable as approaching, even more closely than
the older states, the principles of justice represented in the great Irano-Semitic
imperial tradition. In that tradition, the absolute monarch was expected to
rise above all more limited interest groups and privileged classes and maintain
some balance among them, so that the stronger should not freely override the
weaker. In the first instance, the Arabs won support in Roman territories and
probably in the Iraq and even parts of Iran by curbing a persecuting ecclesias-
tical rule and imposing equality among the sects. 'Umar's organization of the
empire had been designed to maintain the unity and purity of the Muslim
Arabs, but at the same time it confirmed the subjects' hopes in considerable

241

242 THE CIVILIZATION OF THE HIGH CALIPHATE

measure. This may best be illustrated in the crucial matter of taxes. ' Arrange-
ments about taxes varied from locality to locality, depending upon the mode
of conquest and on the local traditions. Some municipalities were permitted to
handle their own tax collecting, handing on a fixed sum to the Arabs. But in
most of the richest lands, the Arabs had maintained a rough supervision of
the taxation. This had been necessary from the start in the Sawad of the Iraq,
so dependent on central administration, and experience there had been gener-
alized. This system proved efficient (under shrewd men of the Quraysh), ten-
ding to bypass superfluous middlemen, and was informed by an intention of
equal justice which 'Umar regarded as inseparable from Islam and which was
often a boon to the weaker subjects.

In the old empires, or at least in the Roman, with time more and more
privileged families and institutions had acquired preferences and exemptions
in tax matters, leaving a disproportionate burden on such as had won fewer or
no privileges. It was precisely the privileged great families who held power in
their hands, and the central government was so far committed to them politi-
cally that its efforts to rectify the imbalances were ineffective. In the Roman
empire the point had seemed to be coming (a point of no return for any state
of the Agrarian Age) when the imperial resources were so compromised by such
internal commitments that they were no longer sufficient to allow a reforming
ruler to overcome his privileged subjects in case of contest. In this situation,
only an overthrow of the privileged class generally, from within or from with-
out, could redress the imbalance. This the Arabs performed. Beholden to no
local elements, they could exercise their own central authority freely. Where
the tax structure was the least fair, they insisted that it be set in balance; for
(and here justice and interest coincided) a balanced taxation yielded a larger
proportion of the revenue to the central authorities.

What had happened found special expression in the new incidence of the
poll tax, the tax to be paid by an individual as such, rather than as holder of
lands. In both the Sasanian and the Roman empires a poll tax had been col-
lected; in each case, the privileged elements had been exempted from it. (In
the Roman empire, the exemption seems to have been extended even to
the city mobs, who probably had more political leverage than their opposite
numbers in the more agrarian-oriented Sasanian state; thus it fell chiefly on
the mass of peasants.) The Arabs now continued much the same system. But
the 'privileged', who were to be exempt from poll tax, were to be only Arabs.
Those who had been privileged under the old empires had to pay the poll tax
— unless, indeed, they converted to Islam and joined the ruling Arab com-
munity (as some did, precisely to maintain their privileged dignity). All the
non-Arab non-Muslim population was classified together under one rubric,
dhimmi, as recipients of Muslim protection provided they submitted to

' Early Muslim taxation and especially the poll tax have been illuminated by Daniel
C. Dennett, Jr., Conversion and the Poll Tax in Early Islam (Harvard University Press,
1950), a keen study which is, however, unfair to Wellhausen in its broader generalizing.

THE ISLAMIC OPPOSITION 243

Muslim rule — however much that dhimmi population differed internally in
degree of social privilege.

The new privileged class was not indeed subject to the caliph in the same
arbitrary manner as had become a monarchical ideal in many minds. Yet at
first its situation provided equivalent assurances. It was kept apart from the
bulk of the population and under a special discipline. Hence at first it must
have weighed relatively little, in day-to-day matters, upon the masses. The
older class structure persisted in ordinary life except that its very top level
was gone in many areas; but at least symbolically (and increasingly in fact)
the remaining privileged groups were levelled before the conquerors, to whose
collective justice appeal could be made. Indeed, all were not contented. By
the time of 'Abd-al-Malik, an increasingly close supervision of taxation to
maximize the yield was producing peasant discontent in some places, renewing
the perennial agrarian tension between the producers and their exploiters. But
in the central parts of the empire, where the new regime ruled most directly,
and whence it derived its major strength for controlling more outlying areas,
the major urban classes of the subject communities seem to have been glad to
support the government.

However, with the gradual breakdown of the isolation of the ruling Arabs
from the general population, the special situation ceased to hold. Hence there
arose pressure on the government to control more closely the new privileged
class. As the relation of privileged and subjects came to have traits more nor-
mal for an agrarian society and as the need for absolute monarchy came, cor-
respondingly, to be more strongly felt, the Marwani caliphate came steadily
closer to the ideal of the absolute monarchy.

Already by the time of 'Abd-al-Malik, the social intermingling had gone far.
On the one hand, many of the conquering Arabs had themselves bought lands
(or been granted former crown lands by the caliph), and become landlords.
The distribution of the revenues was no longer purely a matter of 'Umar's
army di wan ; many Arabs now began to have personal, private contact with the
sources of revenue. On the other hand, non-Arabs had begun to settle in
'Umar's military garrison towns; so that even these towns, for instance Kufah
and Basrah in the Iraq, were becoming ordinary cities such as Damascus had
always been, closely linked with the surrounding countryside. Officials and
grandees became Muslims to maintain their status. Merchants found markets,
landless peasants found work in the Muslim towns; they learned Arabic, often
became Muslims, and settled down alongside the children of the conquerors.

The result was that the administration of the conquered lands and the
governance of the Arab ruling class could no longer be readily separated. It
was no longer possible to deal with the Arabs in one way, as an occupying
military force, and with the conquered populations in another way, as an
occupied territory administered on its own principles, of no direct concern to
the Arabs. Either the Arabs must be gradually submerged into the population
at large, their business increasingly carried on in the languages and according

244 THE CIVILIZATION OF THE HIGH CALIPHATE

Chronology of the Marwani Umayyads, 692-750

692-744 The Marwani Umayyads: The empire continues to expand till

about 740, and internal peace is broken only by Khariji revolts
(increasingly extensive) and occasional scares raised by Shi'is
(' Alid partisans) ; the administration is consolidated and
regularized; a 'pious opposition' to Umayyad rule, of many
shades of opinion, centres at Medina and is increasingly inclined
to support the 'Alid claims raised at Kufah; specific events are:

692-705 'Abd-al-Malik undisputed caliph, Arabizes the administration
(696, Arabic coinage); Hajjaj b. Yusuf at Wasit (694-714), as
his lieutenant in the former Sasanian provinces, bloodily
suppresses dissenting Arab movements, encourages economic
development

705-715 Walid I, caliph, conquest of Spain and Sind, and first conquest
of Transoxania. Succeeded by Sulayman (715-717), who fails
to take Constantinople (717) and permits the 'southern' Arabs
(Kalb and allies) to triumph over the 'northern' Arabs (Qays
and allies, among them Hajjaj 's men), intensifying feuds among
the Arab soldiery

717-720 'Umar II b. 'Abd-al-'Aziz, caliph, whose piety, of the new
Medina type, conciliates even Shi'is and Kharijis; he
encourages admission to the ruling class by conversion, and
attempts an 'Islamic' solution to the problem of taxation on
converts' land. Succeeded by Yazid II, 720-724

724-743 Hisham, last great Syrian Umayyad caliph, organizes the

administration for efficiency; Transoxania is subdued, but the
Shi'is become restless in the Iraq, the Kharijis everywhere.
(Zayd, an 'Alid, revolts at Kufah, 743). Succeeded by Walid II,
743-744. John of Damascus (d. c. 760), major Greek Christian
theologian, associated with the Umayyad court

744 _ 75° The third fitnah civil wars: a dissident Umayyad force led by

Marwan II, destroys Syrian Umayyad power and suppresses
three other rebellions representing groups of the 'pious
opposition' till it is overthrown by a fourth, the 'Abbasi, which
reunites the empire

For comparison in eastern Europe :

717-741 Leo the Isaurian reorganizes the Byzantine empire to resist the

Arabs

For comparison in western Europe :

714-741 Charles Martel restores strength to Frankish kingdom and

defeats an Arab force in northern Gaul

THE ISLAMIC OPPOSITION 245

to the principles of the conquered lands, or the administration of the con-
quered lands must be made a part of the internal government of the Arab
ruling class.

Supported by the social resources of the Arab garrison towns, 'Abd-al-Malik
and his lieutenant in the east, al-Hajjaj b. Yusuf, were able to avoid the for-
mer alternative. First, the military decision which had restored the jama' ah,
the united community, was followed up with a determined power policy de-
signed to maintain central authority in detail over all tendencies to tribal
autonomy. This policy was incarnated in al-Hajjaj. Al-Hajjaj began his rule
in the Iraq (694) with terrifying violence, designed to reaffirm Abd-al-Malik's
control in the garrisons themselves. His first task thereafter was to crush the
rebellious Kharijis of the Iraq and Iran; this was accomplished within six
years. The Khariji pattern may have afforded some satisfaction to the dhimmi
non-Muslim population, possibly especially to the peasantry, because it might
seem to offer a return to the clarity of the Muslim policy of non-involvement
of 'Umar's time; but in the long run it was not so satisfactory as an absolute
monarchy would seem to be. The non-Muslim city populations seem to have
readily accepted 'Abd-al-Malik.

But by 701, al-Hajjaj 's high-handedness had pushed the Iraqi army — led
by a chief from the Qahtan tribal faction in Kufah, Ibn-al-Ash'ath — into a
rebellion in which Basrah and Kufah joined together. When it was put down,
his building of Wasit as new capital, with a Syrian garrison, confirmed the out-
come. Henceforth the Iraqis in Kufah and Basrah were little trusted either as
citizens or even as soldiers for frontier expeditions. This outcome in the Iraq
meant, speaking more generally, that the Arab ruling class there were ceasing
to be rulers as such and were becoming part of the subject population; and
something analogous was happening, if less rapidly, in many other provinces,
where the line between Arab and non-Arab was becoming ever more a matter
of social privileges rather than of political function. Yet with centralization,
Arabism was preserved in a new form.

It was appropriate to the new role of the Arabs that, from this point on,
al-Hajjaj dedicated himself, till his death in 714, to restoring, to the profit of
the treasury, the irrigation works of the Mesopotamian plain. Though the
costly irrigation of the Syrian hinterland (which the Roman empire, with its
interest in Christian holy places as well as in Syrian products for the colder
Anatolian highlands, had patronized) was kept up under the Marwanids, more
effort was put in on irrigation in the Iraqi Sawad and also in the Jazirah. (This
latter area had been neglected, it seems, when it marked a frontier between
Roman and Sasanian empires.) The irrigation of the Mesopotamian basin gave
larger returns on investment than the irrigation of Syria. Thus even while the
political role of the Iraq was being reduced, its economic role was being magni-
fied. We may say that 'Abd-al-Malik had restored the Hijaz-Syria axis which
Mu'awiyah had depended on, and which was an intelligible outgrowth of
Muhammad's enlarged reconstitution of the Meccan mercantile system in the

246 THE CIVILIZATION OF THE HIGH CALIPHATE

Hijaz. But henceforth the former Sasanian lands became increasingly impor-
tant, and with them the agrarian basis of the empire and the pressure for the
empire to be assimilated more firmly to the old Irano-Semitic imperial tradi-
tion: a tendency which was finally consummated after two generations in the
shift of the capital back to the vicinity of the Sasanian capital. 2

On the basis of 'Abd-al-Malik's policy of centralizing power, the central
authority could build its direct bureaucracy. It soon imposed the Arabic
language on the administration, replacing Greek and Pahlavi; the coinage was
minted and the tax books kept henceforth in Arabic. The official classes, at
least on the upper level, were made to learn Arabic, and promotion came to
depend on a man's skill in Arab ways. Thus was assured the supremacy in
civilian life, as in military life, of Arabic-speaking cadres. But this supremacy
implied, at the same time, integrating into a single process both general ad-
ministration and the governance of the Arab class; and this whole process was
dominated by the indigenous administrative tradition, not by Arabism.

Though he was not fully an absolute monarch, 'Abd-al-Malik went far to-
ward establishing, in the course of fighting for power and of confirming it, the
unbounded authority of his own office over the Arab community. This policy,
forged by political necessity, fitted in perfectly with the monarchical expecta-
tions of the conquered subjects. In his relations with the Arabs, 'Abd-al-
Malik stood for the principle of the j ama'ah, the moral and political unity of all
Arabs under the aegis of Islam; a unity which was to be enforced, if necessary,
by military power. But the same principle of unity served as a basis for mak-
ing the financial and agrarian administration of the whole empire more uni-
fied; an endeavour which in turn increased the role of the caliph as master of
an increasingly entrenched central bureaucracy, through which the interests
even of Muslims could be governed.

The new status of Islam and of the Muslim government was given symbolic
form in the visual arts. The new coinage in Arabic was doubtless in part a ges-
ture directed against Byzantium, whose gold denarius coins (in Arabic, dinar)
had formed the dominant monetary standard in Syria and the Hijaz; it was an
assertion that Muslim rule was permanent and fully independent, whatever
temporary tribute-money the governor of Syria had had to deliver during the
fitnah wars, and that it could perform all the functions which the older im-
perial governments performed . The symbols on the coins appropriately pointed
to Muslim power. An experiment was made with a portrait of the caliph, an
absolutist gesture for which the Muslims were not yet ripe; but the successful
symbolism was more abstract and referred to Islam directly. Some coins
showed a mihrdb (niche indicating the direction of Mecca), symbolic of the
common worship of all Muslims, and the Prophet's lance; but finally (perhaps
in response to a Byzantine substitution of the figure of Christ for that of the

2 Oleg Grabar has touched on the importance of such economic shifts and how all
fields of history, such as art and economics, are interrelated. See for example, his 'Islamic
Art and Byzantium', Dumbarton Oaks Papers, 18 (1964), 67-88.

THE ISLAMIC OPPOSITION 247

emperor) the word of God itself was used — in Qur'anic phrases which presaged
the might of God and of the Muslims.

The use of writing as the only imagery on coins was a daring innovation,
which the population nevertheless accepted, showing their confidence in the
new power; more than that, it was an iconographic stroke of genius, taking
advantage of the strong clear lines of the squared-off kufw form of the
Arabic script to produce a design at once highly abstract and very immedi-
ately symbolic. In other public places, a comparable symbolism of Muslim
power was displayed: at the Dome of the Rock, now built in Jerusalem to
assert the continuity of Islam with the pre-Christian prophetic tradition, such
symbols of power as crowns and pictures of holy buildings were used. In line
with an iconophobic tendency already prevalent among Jews and among
those Christians most opposed to Byzantium, and adopted by at least some
Muslims, figural images were avoided in public places. (This tendency was
later to become important in Islamicate art generally.) 3

'Abd-al-Malik built well. After thirteen years, in 705, he left a well-estab-
lished power to his son, al-Walid I. With al-Walid, a hereditary principle of
succession was accepted for the first time without demur; or, more exactly,
the principle that the monarch could dispose of the succession at will within
his family, without more than formal intervention on the part of the Arab
notables — for 'Abd-al-Malik's settlement provided that power should go to his
several sons in turn. It was under al-Walid that the last of the outlying con-
quests occurred, in Spain and Sind. Al-Walid ruled uncontested and ably for
another ten years, continuing his father's policies, till 715, and left the power,
again without contest, to his brother Sulayman. As the Arabs became accus-
tomed to such power, the caliphal state was more and more moulded after the
image of a true absolute monarchy, maintained by a centralized civilian
officialdom. 4

Discontent with the Marwdnids, and the Piety-minded opposition to them

'Abd-al-Malik was able to win considerable favour at Medina — perhaps even
more than had Mu'awiyah — despite the great unpopularity of his regime in
the Iraq. Since 'Uthman's time, there had grown up at Medina a new genera-
tion of men who saw themselves, as heirs of the traditions of Muhammad's
own city, the special custodians of the ideals of Islam, if no longer of its power.
'Abd-al-Malik took a special interest in the religious questions that were
interesting them, and showed his respect for their opinions. If 'Abd-al-Malik

3 In addition to the work by Oleg Grabar cited in note 2 above, see his 'The Umayyad
Dome of the Rock in Jerusalem', Ars Orientalis, 3 (1959), 33-62. We shall deal later
with the iconophobic tendency, already visible here, to avoid figural images.

* Martin Sprengling, 'From Persian to Arabic', Amer. Journal of Semitic Languages,
56 (1939), 175-224, 325-36, is a very valuable study of early Muslim administration,
largely on the basis of the later katib literature. (The latter part of the article, on the term
wazir, has been largely superseded.)

248 THE CIVILIZATION OF THE HIGH CALIPHATE

carried on the political side of Muhammad's heritage, most especially in his
insistence on the jama'ah principle of Muslim unity, he could also claim that
the Qur'an, as represented in its most ardent spokesmen at its home city, still
played the supportive role toward his policies it had played since Muhammad.
But gradually under the Marwanids the political and the ideal sides of Muham-
mad's heritage came not only to be represented by different sets of people, but
these largely found themselves in tacit or even active mutual opposition. The
civil wars had in fact seriously undermined the balance that Muhammad had
established. The defeat of the attempt of the Hijaz Muslims to regain leader-
ship under Ibn-al-Zubayr and the suppression of the Shi'i and Khariji
movements necessarily still rankled.

As the new generation — and not only at Medina — began exploring what
Islam could mean to them personally now that its political triumph was as-
sured, their thinking was deeply coloured by the experiences of the civil wars.
The notion of the jama'ah, the unity of the community, did not suffice as a
comprehensive Islamic ideal, even when it was accepted on the specific point
of who should be caliph. Many came to feel (as the Kharijis had early insisted
in their own way) that the Qur'an should play a more active role in the life of
the community. Thus many of those who had been associated with the de-
feated parties, not only in the Iraq but even in the Hijaz, came to constitute a
semi-political, semi-cultural body of opposition to the ruling trends among
the Arabs. In their view, the community at large which gave allegiance to
Islam seemed mostly devoted in fact to enjoying the fruits of conquest under
the leadership of men whose position in power had resulted largely from force
and from tribal alliances. Given such a mood, it was natural that, among those
relatively few who were taking a special interest in Islamic ideals, an opposi-
tional standpoint, in greater or lesser degree, became common. It was in this
cpirit that they began to develop a more intimate and more universal concep-
tion of Islam. These men envisaged a society which should embody justice on
earth, led by the most pious among the Muslims.

They had before them the example of a few men who, by their own intense
sense of the divine challenge in their personal lives, reminded all Muslims of
what the Qur'anic challenge could mean to them simply as human beings.
From the very first generation, there had been individual Muslims, like Abu-
Dharr (d. c. 652), who took their piety sufficiently seriously to undertake the
rebuke of the early caliphs when they seemed too worldly. But now such
figures were no longer just eccentric purists.

The model Muslim of Marwani times was Hasan al-Basri (d. 728), popular
preacher (and for a time qadi, judge) at Basrah. He was brought up in Medina,
the son of a freedman, in circles close to Muhammad's family. He meditated
on the spirit of the Qur'an intensely and showed its fruits in his upright, fear-
less conduct as well as his overwhelming sermons at Basrah. He won the
respect of all parties, of al-Hajjaj the governor and of those who most hated
al-Hajjaj. Hasan accepted the Marwani rule, but he criticized the Marwanids

THE ISLAMIC OPPOSITION 249

when he thought them wrong. When 'Abd-al-Malik was punishing men in
Syria for raising religious discussions that carried implications of criticism of
the regime, and in particular for suggesting theological positions he disap-
proved of, it came to his attention that the well-known Hasan at Basrah was
teaching just such positions (in this case, that God, being necessarily just,
would necessarily leave human beings free to choose to do right if they wished).
'Abd-al-Malik asked for clarification from Hasan, evidently expecting a pru-
dent disavowal. But Hasan, who felt it necessary to emphasize people's free-
dom to do right, so as to bring home their moral responsibility, responded with
a forthright justification of his position; and 'Abd-al-Malik let him be. 5

Often those inspired by Hasan's example and teaching, even if they did not
directly oppose Marwani rule, encouraged an anti-government spirit. Grad-
ually these and others developed a generalized critique of the Marwanid
caliphs and of their policies, as well as of the Arab life they represented. In
the course of such discussions, Islamic religion itself in the full sense, as a com-
prehensive aspect of human culture, began to take form. As long as they
were isolated, however, these pious dreamers remained politically irrelevant.

But such men found allies among others who were less religiously minded.
The ordinary Muslim Arab supported the Marwanid rulers in Syria as religious
leaders, inasmuch as, by maintaining Islam, they guaranteed the moral unity
of the Arab community — the unity of a ruling class that had to maintain its
position in a subjugated land of non-Arabs. But there were many groups that
had material objections to the policies of the Marwani regime, despite its
strong moral position. Built on the power of Syrian Arab troops as it was, it
favoured the Syrian over the other Arabs, over the rival Arabs of the Iraq,
and above all over the wealthy old families associated with Muhammad and
centred on Medina. The bloody severities of al-Hajjaj against the proud in-
dependence of the great families were never forgotten. These politically less
favoured Arabs would have liked to gain more political leverage or even to
change the locus of power to some point nearer themselves, in this case pre-
ferably setting up as rulers one of the old Medina families which would depend
on support from the Hijaz and the Iraq. Such discontents were complicated
and reinforced when, after the time of al-Walid, the Marwanids themselves
seemed to take sides in the Arab tribal factional quarrels.

In the long run more important than these discontented Arabs was a class
of men whom the society of the garrison towns was creating but whose interests
were in conflict with that society. A rising number of non-Arabs had aligned
themselves with the Arabs and joined generally in the life of the Muslim Arab
centres. They had become duly affiliated as 'clients' to an Arab tribe (and
hence were called Mawali, 'affiliates' or 'freedmen') and had learned to speak
Arabic; and many had embraced Islam, as essential to full identification with
the Arab community. Yet even when they became Muslims, neither they nor

5 On Hasan Basri it is worthwhile comparing the studies in Der Islam by H. Ritter
(21 [1933], 1-83) and H. Schaeder (14 [1924], 1-66).

250 THE CIVILIZATION OF THE HIGH CALIPHATE

their descendants were treated as full partners in the Arab privileges; indeed,
they could hardly be so treated without breaking down the social order formed
by the garrison towns, which presupposed a reasonably small elite tied to-
gether by common traditions and experiences. The Muslim Mawali, however,
could take little comfort in this; they had every reason to wish for a govern-
ment and a society which would distribute its favours on the basis of accep-
tance of Islam rather than on the basis of Arab descent. Thus large groups of
Muslims, non-Arabs as well as Arabs, wished for some change and might, if
any weakness should appear in the Marwani regime, prove receptive to
projects of revolt; always provided they could hope that an overturn of the
accepted order would be in their favour.

But no rivals to the Marwanis could present a plausible case unless the
moral position of the Marwanis, based on the jama' ah ideal of unity, could be
undermined. It was here that the oppositionally inclined among the Piety-
minded could play a role and at the same time escape their political isolation.
Such men and women, for whom Islamic piety took precedence over any other
interest, were found among both the disaffected Arabs and the Muslim Mawali,
and even among the Syrian Arabs themselves. As they took leisure, now that
the great excitements of the days of first conquest were over, to consider more
closely what 'Islam' ought really to mean, they realized it must reach beyond
simply being the envied badge of a favoured ruling class. The grievances of
their fellows made them see more keenly what was wanted. But such a
viewpoint led them to discover the points of moral weakness in the regime;
in doing so they became rallying points for all the elements of potential op-
position. The outcome was that at last they gathered it into an all-Islamic
tendency, rather than merely a series of local rebelliousnesses that might have
broken up the community and so, probably, done away with Islam altogether.
But they would not have had their great influence had there not been other
motives for opposition present among large numbers.

I refer here to the 'Piety-minded' element as a general term to cover all the
shifting groups opposed to Marwani rule, or at least critical of current Muslim
life, so far as their opposition embodied itself in idealistic religious attitudes. I
am speaking, of course, primarily of the religious specialists, later called
'ulama' , who provided much of the leadership. At the same time, the more
pious of their followers are to be included, for there was no sharp line, at first,
between 'ulama' scholars and others. Only gradually did the social element
that we designate as the 'Piety-minded' resolve itself later into sharply dif-
ferentiated Sunni and Shi'i 'ulama', followed with lesser or greater sectarian
devotion by partisan groups in the wider population.

The position of the Piety-minded opened the way to very extensive cultural
implications, as we shall see. The Piety-minded rarely objected to the fact of
Islam's being the badge of a ruling class; they were content to leave the Chris-
tians and Jews peacefully in their state of subjection. But at least within the
Muslim class they wanted Islam to be more than merely such a badge, as it

THE ISLAMIC OPPOSITION 251

often was in Marwani times; to be more than a minimum standard of public
conformity in point of key beliefs, standard ritual, and elementary morality.
Without formulating this in any abstract way, they expected Islam to carry
with it its own law, its own learning, its own etiquette, its own principles of
private life and of public order: to be self-sufficient without any reference (in
principle) to pre-Islamic ways as such. At that time, they made no specially
radical demands on any aspect of Muslim life: grievances were relatively few
and could be pinpointed. Yet the tenor of their thought was that the new
society should be freed of the corruptions of the old.

This outlook had historical roots deeper than merely the play of interests
that followed the civil wars. The very origin of Islam gave the Piety-minded
a point of leverage for associating their ideals with the interests of the dis-
affected. The Islamic tradition already under Muhammad had developed poli-
tical implications, and this commitment was confirmed in the Riddah wars.
Hence it already carried a responsible and egalitarian social commitment. In
Marwani times, this commitment could be turned against the Marwani power
on the basis of Islam itself, interpreted to justify the opposition to Syrian
Marwani power. The concerned could count on the non-Syrian Arabs' harking
back to a pre-Marwani Islamic age supposed to have been more legitimate
and just. At first, this memory referred generally to the early Muslim society
from Muhammad on, and notably to the time of 'Umar; there they had the
concrete example of a pious society, the details of which were not yet forgotten.
Gradually, at least among Shi'is, the ideal period was limited to that of
Muhammad himself, except possibly for 'All's own reign. In either case, the
concerned could rally to such a memory all the political strength of the
various oppositional elements.

The Marwanids were by and large neither more nor less pious than most of the
followers of these Piety-minded men who opposed them. But their strength
was based on a sense of the power and unity of the Arab ruling community
at large and on the special position of the Syrian Arabs; they were not in a
position to espouse a minority's impractical programme. They perforce made
some use of the administrative machinery, the economic order, the legal stan-
dards, the arts, and the learned heritage of the peoples in whose midst they
found themselves. Their palaces were decorated in the usual Hellenistic fash-
ion, the taxes they raised were essentially the same taxes as those raised by
the governments before them, and if their records were — after a time — kept
in Arabic instead of in Greek or Pahlavi, it was nonetheless after the Greek or
Pahlavi manner. What else could they do? There did not exist even a pagan
Arabian pattern for doing these things, let alone a purely Islamic one. Never-
theless, because of their un-Islamic 'innovations', they were labelled impious
by the opposition and sometimes accused of betraying Islam itself; the task
of working out the social implications of Islam was to be carried out as much
as possible without them.

252 THE CIVILIZATION OF THE HIGH CALIPHATE

The programme of the Piety-minded

The proposals that the Piety-minded offered in response to the discontents pre-
sent in the social and historical situation were summed up as government
according to 'Qur'an and sunnak' (custom) . But what could this mean in parti-
cular? What was the Qur'an was clear enough, but in itself it did not provide
for day-to-day cases. The crux lay in defining the sunnah, which as a word
merely meant the 'established practice'. What was objected to as contrary to
sunnah was the seemingly arbitrary departure from what Muslim Arabs had
expected — or hoped for. The restrictions and indignities for the privileged
Arab families which were inseparable from the development of a centralized
monarchy were seen as innovations, called bid' ah; and the seemingly more
liberal days of earlier rulers — especially of the Medina caliphs and of Muham-
mad himself — were recalled as models of what all could agree ought to be: as
sunnah (the word had been used for approved customary practice by the
Arabs before Islam). At the same time, it was recognized that the bid' ah, the
deplored innovation, was not entirely a matter of the rulers; their power and
arbitrariness were partly the consequence of the moral laxity and luxurious
habits of the Muslims themselves — for it was in these terms that moralizers
naturally saw the assimilation of the Arab ruling class into the cultural and
social life of the occupied lands. Accordingly, abiding by the sunnah would
mean restoration, for both rulers and Muslims at large, of the norms of the
primitive caliphate and (or, among many Shi'is, only) of Muhammad's time;
what did not go back to such times was bid'ah and ought to be eliminated
from Muslim life.

This programme (if so generalized a demand can be called that) rested on
the assumption that the Muslims as a class were to retain a distinctive char-
acter, keeping separate from the subjected peoples and all their diverse ways
so far as possible. If a member of the subjected peoples embraced Islam —
which as a human being he had the privilege of doing — he should assimilate
to this Muslim character. But the distinctive character of the Muslims could
not be simply their Arabism. They already felt themselves Arabs, and any
distinction of worse or better Arab was expressed in divisive tribal rivalries.
The distinctive character of the Muslims must be their Islam. Proper sunnah
practice, then, should not be primarily Arab sunnah but Islamic sunnah.
What was to be looked for in the times of the primitive caliphate was its repre-
sentation not of Arab norms but of Islamic norms.

This meant a reversal of the place of Islam in the Arab society in the con-
quered lands. From being a society o'f Arabs who happened to be bound to-
gether by Islam, it must become a society of Muslims who happened to use
the Arabic tongue and respect parts of the Arab heritage. This meant, ulti-
mately, that Islam was not to be an Ishmaelism, analogous to the Israelism
of the Jews, in which converts could enter fully into the community only as
they were assimilated into a stock all of which, in principle, descended from

THE ISLAMIC OPPOSITION 253

Abraham. Despite the obvious possibilities in exalting the Arabs as being of
the line of Ishmael and producing an ethnically-bound community, the com-
prehensiveness of the Islamic vision prevailed, in conjunction with its sense of
political mission, to mould Islamic idealism in a more universalistic direction.

This did not always mean a radical reversal; much that was old- Arabian
had been sanctified as Islamic by being accepted in primitive Islamic times.
In some points there was convergence: for instance, the sense of the invio-
lability of the Arab tribesman was undermined, indeed, by his subjection to
Islamic penal sanctions; but it was otherwise reinforced by the fact that a
Muslim had an individual dignity as such that no other Muslim could be justi-
fied in abridging: all Muslims were to be as one tribe. The Piety-minded ten-
ded, therefore, to support what was otherwise a tribal demand: that every
free Muslim should be accorded that personal liberty and dignity which was
expected by the Arabian tribesman — being bound to obey no man without
his own assent. Consistent with this anti-state attitude was a general distrust
of the more elaborate forms of urban luxury and social distinction. The old
Arabs, far from the centres of wealth concentration, had had little in the way
of visual arts, for instance. Men like 'Umar had freely enjoyed the art objects
that they had inherited by conquest; but now some of the Piety-minded raised
religious scruples against such things. Probably encouraged by a distrust of
luxury already present in the populistic monotheisms, they likened the use
of precious carved and figured objects to idolatry and condemned it as
innovation.

A most important consequence of the new attitude, which gradually be-
came clearer, was that all free Muslims ought to be treated on an essentially
equal basis. Not that the great families of Medina, descended from Muham-
mad's closest associates, should not be accorded social priority; but that in
point of public policy all should be on a level. Here a logical consequence of
the Piety-minded viewpoint accorded with the widespread demand, put forth
by most of the discontented, that there be no arbitrary distinctions of rank or
class within the Muslim class; neither distinctions based on tribal allegiance
(which led to the disruptive factional feuds), nor distinctions based on mili-
tary power blocs such as that of Syria on which Marwani power rested. The
demand was that at least all the Arab Muslims, of whatever bloc, should share
equally in the fruits of conquest (at least so far as they had participated in it)
— in the booty and the pensions based on revenue from the conquered lands.
Nor (added some) should Muslims recruited from the subject populations be
discriminated against: they and their descendants should have the same
rights, obligations, and liberties as those descended from the conquering fami-
lies. (This latter was a corollary only slowly accepted by a great many of the
Arabs, and never without reservations. Various minor privileges for Arabs
found their way into the Shari'ah law.) This popular demand, in its various
aspects, struck sharply at the bases of Marwani rule and even at the require-
ments of their everyday administration. But obnoxious as was this demand to

254 THE CIVILIZATION OF THE HIGH CALIPHATE

the principles of empire, it fitted in very well with the aspirations of the op-
position; and it was required, moreover, to assure the homogeneity of Muslim
life-ways which the demands of the Piety-minded looked to. In particular,
assimilation of converts could not be complete without it.

But the Piety-minded could not rest content with a few general principles.
The expectations of the Piety-minded called for settling, in explicit detail,
what ways should be associated with Islam. Though the full scope of the task
only gradually became apparent, it was launched in Marwani times. It was
the work of several types of specialists — the intellectuals of the early Muslim
community. The earliest tradition of such specialization was that of the
Qur'an-reciters, dating from the life of the Prophet himself; they had been
prominent in the early Muslim disputes. Now the Qur'an-reciters began to
develop a complex intellectual discipline, in which there came to be many
schools each with its own set of minor variant readings; eventually they made
of declaiming the Qur'an an elaborate fine art. It was apparently in connec-
tion with Qur'an-recitation that the analysis of Arabic grammar began in the
Iraq, so as to ensure correct parsing.

Perhaps less central than preserving the Qur'an itself, but in the long run
more fruitful of varied cultural implications, was the collection of lore about
the prophets and especially about Muhammad. People would share anecdotes
about what he had been seen to do or say, called 'news', hadith, and gradually
an increasing number of such hadith reports came into general circulation:
reports about what the Prophet and his associates had said or done that would
be relevant to modelling the pious life. 6 Probably some of the associates of
Muhammad had already in their day been more or less well known as reporters
of his doings; they would have found an audience in the pious, the patriotic, or
simply the curious. In the following generations, hadith-reporters came to
share with Qur'an-reciters the repute of authority in matters religious.

Among such men were some who systematized these reports sometimes
into full-scale narratives. A school which traced itself to Muhammad's cousin,
'Abdallah Ibn-' Abbas, attempted to find and transmit the occasion for as
many as possible of the passages in the Qur'an and to explain, more generally,
what was meant by the various passages. Others collected narratives of
Muhammad's campaigns or other events. Muhammad Ibn-Ishaq of the Mawali
of Medina (d. 767) composed a detailed life (sirah) of the Prophet, which in-
cluded a history of monotheism in Arabia and of the Quraysh before Muham-
mad's time, and full reports on the Muslims of his time. A major theme of the
history was the honour of the Ansar, Muhammad's Medinese associates, over

6 It has been common usage to translate hadith by the word 'tradition', but a more
mischievous case of mistranslation would be hard to find. In many contexts, and often in
crucial ones, hadith means just the opposite of what is implied in the Knglish word
'tradition'. The hadith report was documented, not anonymous; it was explicit and
written, not oral, immemorial, and imprecise; it was very often just contrary to custom
as practiced. Compare the discussion in the section on usage in Islamics studies in the
Introduction.

THE ISLAMIC OPPOSITION 255

against those Meccans who had remained hostile till the last minute. The
implication was that (despite the principle that acceptance of Islam wipes out
the past) the later Medinese, descendants of the Ansar, had more right to
leadership among the Muslims than had the Umayyad Marwanids whose an-
cestors had opposed Muhammad at Mecca. (Ibn-Ishaq's own preferred candi-
dates for office seem to have been the 'Alids.) Ibn-Ishaq was part of an inci-
pient Islamic school of history which was not limited to the prophets but ex-
tended to the Arab conquests and to all the events of the early community,
preserved anecdotally in the form of hadith reports. Thus the Muslims main-
tained a sense of historical identity which was coloured by opposition to the
dynasty in the name of Islam.

But still more immediately relevant to social purposes than the specialists
in Qur'an-recitation and in hadith-reporting were the men who specialized in
the attempt to state, in precise legal terms, what the Islamic way of doing
things should be. In Basrah and Kufah, in Medina, in Damascus, they took
current practice in the Muslim courts as a starting point and refined it. The
Muslim courts, designed chiefly to settle disputes among the occupying
soldiery, were still relatively unsophisticated, lacking an established body of
technique and limited in the range of their cases; they afforded an excellent
opportunity for developing new ideas.

This attempt to determine the proper answer to questions of legal (and
personal) practice was called fiqh, 'understanding'. Qur'anic passages would
be applied when they were clear, of course; but by and large the men of fiqh
had to work on less explicit bases. Sometimes they appealed to a general sense
of equity or of social utility; sometimes to local precedent — the decisions of
respected earlier Muslims in their own centres. Gradually it became the cus-
tom to trace the local tradition (sunnah) back to associates of the Prophet
who had settled locally, so tying it to the sources of Islam. Sometimes the
anecdotes of hadith-reporters — about the Prophet himself or about his associ-
ates, including the first caliphs — would prove relevant to legal questions; well-
known reports of the sort would be used by the men of fiqh. Some of the men
of legal fiqh were practicing qadis (judges). Others remained private experts,
like Abu-Hanifah (699-767), a well-to-do merchant of the Mawali of Kufah
(his grandfather had been an Iranian war captive who was freed and became a
merchant). Abu-Hanifah stood out as a teacher, sharpening the legal rea-
soning of his predecessors and attempting to present a consistent total system
of law. Through the work of Abu-Hanifah, the legal fiqh tradition of the Iraq
became a model of Muslim legal acumen for the time.

All these fields of inquiry, and especially memorized retention of Qur'an
and hadith, constituted 'ilm, 'knowledge' — that is, knowledge of what was
right. Collecting, elaborating, and transmitting such 'ilm was a primary acti-
vity of the Piety-minded. In 'Abbasi times, these lines of inquiry developed
into an elaborate intellectual culture among the Shari'ah-minded 'ulama' scho-
lars. But though many of the Piety-minded were content to work privately,

256 THE CIVILIZATION OF THE HIGH CALIPHATE

educating disciples in their views, private 'ilm by itself was not enough; it
needed to be agreed on by the community and put into practice. This had
political consequences. The unwillingness or inability of the Marwanids to
take the lead here intensified the feeling of discontent with them as rulers. In
a military ruling class, reform was above all public reform and power was
military power. Hence there was always a presumption anyway that serious
change would come by an armed replacement of the top command. There were
no autonomous establishments to reform, short of the state power itself; and
insubordination at that level meant revolt.

On this, many could agree. But revolt required a candidate to replace the
Marwanids; and the candidate himself ought to be a suitable expression of
Islamic piety. At this point the Piety-minded disagreed among themselves.

The problem of authority: 'ilm and imdmate

It was against this background that the disputes proceeded among the Piety-
minded as to what sort of authority, when such was needed, should take the
place which had been held by Muhammad and his immediate successors. Such
authority was needed particularly for practical command of the community,
in assigning tasks such as those of defence or of maintaining public order. But
at least as important, someone was needed who could settle disputes about
'ilm if, for instance, what was claimed as hadith should vary. The Marwani
power, based on the jama' ah, had at its disposal Syrian troops; what sort of
authority would be able to replace it with power based, as the more militant
of the opposition agreed it should be, on 'Qur'an and sunnah'?

In Muhammad's lifetime, the authoritative commander, later called the
imam, leader (for his first duty was to lead in the worship), had without ques-
tion been Muhammad himself. Upon his death, authority had lain, in a general
way, with those closest to him, who knew his ways the best (and hence knew
God's will the best), and were accepted as worthy by the community at large
on the basis of personal acquaintance. The Piety-minded wanted the imam to
be of this sort: in their terms, this meant that he who is the admittedly most
pious and most knowledgeable of God's will ought to have the command, to
be imam of the community; as to whom this implied in practice, they differed.

Perhaps the earliest group to pose this problem in its essentials — already in
the days of Mu'awiyah — were the Kharijis. They broke off from 'All's party
when he appeared to be willing to compromise with Mu'awiyah, and for gene-
rations they fought almost every Muslim government that appeared. Their
solution was an extreme one: it was up to each believer to decide who was the
most pious, or at least an adequately pious commander; to join with others
who acknowledged that man's command; and to separate from any self-
styled Muslims who irresponsibly accepted as imam an unworthy man. The
man who showed by his sinning that he was no true Muslim was unworthy of
rule over Muslims, and any who accepted that rule showed themselves traitors

a
+->

E

'£.

H

4)
43

O

4)

s

43
+j

O

u "

r-* •— !

CI

S3 4)

>

O „

i-i

43

in

W
S3

'3

H

43

d

tH

«•

d

ffl

>.

i— «

.2

'3

<D

Oh

CO

<u

■d

a

d

43

a

d

O

>+3

^3

<3

«

03
43

d

[-1

O

4J

Vh

[-<

m

03

^

>

43

d

o

1

so

<d

a

*3

+j

£

'io

fi

S

TJ

S3

d

'*-]

<

V3

S

_co

■fl)

42

o

tf

43

d

£

u

<=J

&

»

t-c

o

C3
o

•a

'a;

a

42

0)

3

lH

4*

+J

d

tH

43-

>
o

.g

CO

j>>

V*

u

<d

d

43

ot

w

W3
03

o

<u

03

£

(3

'■+-»

-p

'§

o

»H

fk

d

< S3
. d

<d J3

-^ a

o ui .5 £ 42

rt - b " 8 -a

2"^

d 03
13 'G

* o 2

O.SE
<l><2 rt S

b o _■

IT 43 «>g

™T5

r! «

•Sje o>

a >..s

"" c .3
2-52

bo

c
'C

3
43 *0

d n
o « d

° S a

^ a-a

£42 s

BB-a 1 ,

4i o -P

w ri 1-

•o E no

d +• • -

-.a I

d ,« t^
w « d

03 ;^ E3

(/)H rt

^"3 S,

-"24*

•350

«

g<42 £ rt42
o'S*ft H

t»lM)5?S42

J3 43 rt

.«=<d

O - ^^-^

w.Stx, rt
a> " _^> i-i >-o

ss^

S'

43.-^ rt _=

o -a ><a
43 cs

u, K P d

CD

T3 „ cfl
"㤠42
||fe 43

Q

.pQ^

^ <t> .

•h »n "

1 3 ^.a ^i^

! <1 d «oj "S "d
5 'd J5 ri S< —

5t

O u 1 ^
60^3.2

3+50
cue

4) g g

■s 8 a

u 5 c
43 2 8

a

3

03

42
cP

O X)

■■s<S

lT o
0) -*-»

S »

J) a)

£ >
213

43 en

«- s
£5

d 5-
JS ?

C O

d ja
a?

sa
843

o

43 "3
^f

a *>

« s

d 43

.a (u

a^

1-2

^^ o
a ^3

u "

3 P
d >>
o d

■2*

So

G 03

$ a

PI

258 THE CIVILIZATION OF THE HIGH CALIPHATE

to the Islamic ideal and hence no true Muslims either. In effect, the Kharijis —
or the more extreme of them — claimed that their own purist war bands were
the only adherents of true Islam. The rebelling Khariji groups were small
enough and dedicated enough to reproduce and indeed exaggerate the devoted
homogeneity of Medina. The command depended very directly on the assent
of all the faithful to the piety and knowledgeability of the commander, who
could be and was deposed for the slightest of sins. So long as the majority of
the population — being dhimmi non-Muslims anyway — were neutral, the
Kharijis treated them favourably, attacking only rival Muslims.

The Khariji solution had the merit of consistency. But, as we have sug-
gested, the main body of the Piety-minded opposition sought a solution along
very different lines, for they could not afford to abandon the greater part of
Marwani society. Yet, while the Khariji party was eventually isolated from
the main body of Muslims, for some time many who , sympathized with
the Khariji war bands continued to reside in places like Basrah and taught
Khariji doctrines on a modified basis. These men participated actively in the
development of the thinking of the Piety-minded as a group.

The old families at Medina had at first their own solution, supporting for
the imamate one of the families close to Muhammad on the basis of public
recognition in Medina itself. They demanded that the rest of the Muslims, by
supporting such a candidate, renew their allegiance to the descendants of the
mother community in Medina. But after the failure of the attempt of Ibn-al-
Zubayr to retain the caliphate in the Hijaz after the death of Mu'awiyah, this
approach lost ground. Many might have preferred the family of 'Umar, whose
son had great repute at Medina in matters of 'ilm knowledge; many of the
Piety-minded there traced their teachings to him. These tended to accept the
Marwanids in the name of jama' ah unity — and their religious teachings were
received also in Syria. There were those who looked to a reform of the Marwani
regime itself, and in the caliph 'Umar II (717-720) these even saw their man
on the throne for a time. But some of the old families of Medina came to give
their support to one strand in another movement, which in its various forms
became most prominent among the oppositional Piety-minded: the Shi'i move-
ment.

The party of 'All, which was the nucleus of the Shi'ah, had in Mu'awiyah's
day been a fairly small group centred at Kufah in the Iraq. They clung to the
candidature of 'All and of his family for leadership after most of the Muslims
had united in support of Mu'awiyah as restorer of Muslim unity and power.
'All's family soon became for the Kufans representative of the time when
Kufah was near to being the capital of Islamdom, much as the family of
'Abd-al-'Aziz b. Marwan — governor of Egypt under 'Abd-al-Malik — came to
represent Egyptian autonomy and were looked to for leadership by later
Egyptians. But Kufah was, with Basrah, the most important political, econo-
mic, and intellectual centre not only for the Iraq but for the whole eastern,
formerly Sasanian portion of the empire; its cause could easily become the

THE ISLAMIC OPPOSITION 259

cause of all the Muslims east of Syria. (Indeed, it was at the natural centre of
gravity for any agrarian empire in the whole region.) To the extent that 'All
had stood for 'soldiers' rights' and justice against the central authorities, Ku-
fah's loyalties to his house could readily be given a broader interpretation
which deepened their claims in Kufah itself beyond a merely political slogan;
and by the same token, the natural sympathy for their cause elsewhere could
be reinforced with broader moral considerations.

From small beginnings, the Shi'is gradually gained widespread importance,
not only in a town like Qum, an Arab garrison town in the west Iranian high-
lands of 'Iraq 'Ajami, but also in places not normally dependent on Kufan
leadership. The party grew partly through the zeal which 'All's devotees dis-
played, perhaps, and certainly by taking advantage of the historical accident
that not only was 'All a member of Muhammad's family, the house of Hashim,
but the senior lines of 'All's descendants were also the sole progeny of Muham-
mad, through his daughter Fatimah. (Descent through a female was regarded
as secondary, among the Arabs, but was not ignored.) At any rate, an 'Alid
candidature came to be accepted by many, by the last Marwani decades, even
in Medina, which had decisively rejected it at the time of the second fitnah.

There were numerous 'Alid and related candidates and the claims they
made with regard to solving the problem of authority for the godly community
were various. Shi'ism in this period meant support of 'Alid candidates (or
candidates of lines closely related to 'All's) for the command; in itself it im-
plied no particular religious doctrines. The claims amounted sometimes to
little more than a variation of the old Medina belief that the command of
members of the Medina families associated with Muhammad, sanctioned by
the present generation at Medina, would assure a maintenance of Muham-
mad's practice and a restoration of the primitive Medina purity. This was the
normal approach among those Medinese who, preferring, on the whole, the
'ilm knowledge of the tradition of Ibn-' Abbas (the Prophet's cousin) to that of
Ibn-'Umar, also were inclined to wish for an 'Alid candidacy, even, if neces-
sary, at the expense of the jama' ah unity principle of the Marwanids.

But even in the case of a simple Medina candidate, support for the 'Alids
came also from a wider circle. In this wider circle of followers of the 'Alids
there were greater expectations than these mild Medinese hopes. Most com-
monly it was expected that in some way the 'ilm, the knowledge of the ways
of Muhammad, and hence of God's will, was retained more fully and uncor-
ruptedly in 'All's family, or in some branches of it, than elsewhere. This might
be the case merely through close family association with the Prophet, or per-
haps by a more explicit divine intervention. At any rate it was supposed that
with the purer 'ilm of the 'Alids would come a greater right to command; a
right which was only partly the function of the inherited 'ilm, and was more
crucially a divinely sanctioned authority to decide any question of 'ilm which
was disputed. Hence Muslims could look to an era of justice based on true
Islamic 'ilm if 'Alids came to power.

260 THE CIVILIZATION OF THE HIGH CALIPHATE

With the stress on knowledge of the Prophet as basis for true Islamic life,
the close association of 'Ali with Muhammad became more important; and as
this came largely of their family connection, such connection took on new
religious aura. The 'Alids — especially those descending from Fatimah — came
to be called AM al-Bayt, 'people of the house' (an old tribal term referring to
the family from whom chiefs were chosen, often the custodians of the tribe's
sacred objects); here designating Muhammad's more intimate family, which
included 'Alt. But it was recognized also that all the house of Hashim, the
Hashimids, were especially close in relationship to Muhammad; these in-
cluded the descendants of all Muhammad's uncles; not only the Talibids, des-
cendants of 'Ali and his brothers, but also the 'Abbasids, descendants of
'Abbas, who had failed to accept Islam till long after the Hijrah. All these
family lines could be beneficiaries of 'Alid loyalism (a term I use for loyalty to
the 'Alid family in particular but more generally for the whole complex of
attitudes associated with that loyalty) .

By the last Marwani decades there had developed a radical form of the
'Arid-loyalist notion: some people thought that the all-essential 'ilm, and
more generally the inherent authority to decide points of conscience and bear
command over the community, was retained in just one or another particular
line of 'Alids alone. In such a line there would always be one 'Alid who was
inherently the sole legitimate ruler, the legitimate imam, appointed explicitly
(by nass) by his predecessor. Thus there was always in existence a true imam,
whether or not he was at the moment ruling or even making an attempt to
gain rule. It was the responsibility of every Muslim to find him and abide by
his rulings. The pious person might, then, endure the unjust government of the
Marwani caliphs for the time being; for in guiding his own life, at least, he
could refer to the 'ilm of the true imam, which was available here and now.
Such a notion of the imamate made possible a continuing dissident body of
people attached to a continuing line of imams regardless of the fate of parti-
cular political movements. It also encouraged a systematic development of
special religious ideas which could gain acceptance among such dissident
bodies without competing for the attention of all Muslims generally. Here we
have the roots of a sectarian Shi'ism, which later gained major importance.

By the middle of the eighth century there were two such lines of imams re-
cognized by different groups of Shi'is, each tracing its imamate back uninter-
ruptedly to 'All. (The chart showing the candidates of the primitive Shi'ah
shows how this type of imamate became more important with time.) One line
was represented by Ja'far al-Sadiq (d. 765), great-grandson of 'All's son
Husayn; his imamate was to become the basis for most subsequent Shi'ism.
Ja'far's father, Muhammad al-Baqir (d. c. 737), may already have accepted
followers who regarded him as sole legitimate imam of the time, appointed by
nass designation; at least he seems to have been a reporter of hadith. Ja'far
was certainly regarded as such an imam; but he taught hadith, and probably
legal fiqh too, not only to his own 'shi'ah', his special followers, but to others

•a

■a

u

H

en en .£

ft CO

^ CO

co O

X> O

a eU

IS

ft e

B «
S -d

CO >

<* '33

I s

n
o

'3 co
o 53

(ti
Vn 53

CD ft

§•& s

<ni 4-> rt
,g cJ d
CD d T3

s.g.s

US'"

■~ £ o
„-t3 49

« °

^3 CD

£ TJ

«3
"R ^ °

D 5 m

ja .-a &

o bo-r?

O _cu te

262 THE CIVILIZATION OF THE HIGH CALIPHATE

who became prominent later as authorities among the Sunnis. A number of
the most active theological thinkers of the time were associated with him and
with his son, the imam Musa al-Kazim. The authority of the second 'Alid line,
that of 'All's son Ibn-al-Hanafiyyah, seems to have been claimed for at least
three different men, two of whom led revolts at the time of the fall of the
Syrian Marwanids. 7

All these disputes about the imamate among Kharijis and Shi'is and their
rivals gave occasion to, or at least sharpened, further questions in the realm of
'ilm knowledge. For it must be a matter of 'ilm, to know how the commander
of the faithful should be chosen; to recognize who it was could represent Islam,
who it was had 'ilm. Indeed, not only in the case of the ruler, but more gener-
ally as concerned any Muslim, it was of first importance to know what it meant
to be a true Muslim, a person faithful to the divine summons. And this meant
speculation about the nature of the soul, of faith, of 'ilm itself. This specula-
tion was in part enshrined in hadith reports, transmitted by men who con-
vinced themselves that this or that must have been Muhammad's attitude or
that of his associates; in part it took the form of independent argument. As
can be seen from the chart of Piety-minded groups in Marwani times, the
Piety-minded tended to be divided in each chief town into two or more local
schools of thought, which often figured as political factions (sometimes an-
swering partly to tribal factions), each of which developed distinctive religious
teachings; none of these local schools achieved general recognition, though
several found some adherents in several centres besides their main home. 8

Perhaps partly inspired by discussions with Christians, but perhaps still
more inspired by disputes about who was a good Muslim, there were those who
discovered that there was a logical difficulty in affirming at once God's omni-
potence and humans' responsibility for their own acts. Put a bit vulgarly: if
God can do whatever He wills, how can humans deserve either praise or blame
for acts that it is in God's power to force them to do or to omit? More precisely :
if God is truly the sole creator, as the Qur'an seems to imply, He must be not
only more powerful than anyone else but alone responsible for all that is; but
if God is omnipotent in this radical sense, then He must be responsible also for

7 For an analysis of the development of religious and sectarian dimensions in the
early Shi'ah, see my 'How Did the Early Shi'a Become Sectarian?' Journal of the
American Oriental Society, 75 (1955), 1-13. which also discusses the ideas of the early
speculative thinkers, the Ghulat. That article traces the rising fortunes of the early
Shi'ah and in particular of the imamate of the line of Ja'far, but it does not adequately
account for how the Shi'ah could have become so important. It must be remembered
that the early dating, in that article, of certain ideas rests on the overall weight of very
scanty evidence rather than on any sure ground point by point. I would suspect, now,
that the disciplining of the notions of the Ghulat begun by Ja'far came chiefly in
response to the anti-Shi'} reaction under al-Mansur.

s Charles Pellat, Le milieu bafrien et la formation de Gdkig (Paris, 1953), gives a great
deal of information — among other things — about the religious currents in early Basrah.
Unfortunately, Pellat is disappointing in his notions of religion, notably in assuming
naive correlations between doctrine and morals, and in retrojecting later notions like
'SunnI' into the early period.

THE ISLAMIC OPPOSITION 263

Piety-Minded Groups in Marwa.ni Times

kOfah
Shi' ah: looking to 'Alid and related houses for imam (and rejecting 'Uthman):
Moderate (i.e., not recognizing imamate by nass designation)' — often

concentrated on collecting hadith reports (and some were later accepted
by the Hadith folk)
Radical (recognizing one or another line of imams designated by nass — and
rejecting the Medina caliphs as illegitimate) — among these appeared the
speculations of the Ghulat about soul and inspiration and imamate; had
Mawali connections
Murji'ah: willing to accept Marwanid as imams, but reserving right of
criticism ; among these arose the chief school of Kufan fiqh (Abu-Hanifah,
etc.) ; often looked to Ibn-'Abbas in Qur'an interpretation (tafsir)
Kharijis (Ibadis) : a very few survived in Kufah

BASRAH

'Uthmaniyyah: rejecting 'All in favor of 'Uthman, but willing to accept the
Marwanids only for the sake of jama'ah. These overlapped with the school
of Hasan al-Basri, who accepted the Marwanids but asserted the duty to
criticize and oppose them when they were wrong, and who taught the justice
of God, human freedom to do right, and that great sinners lacked true faith

Mu'tazilah: refusing to judge between 'Ali and 'Uthman; took off from Hasan
al-Basri, adding emphasis on unity of God in the sense of a rejection of any
anthropomorphism ; their viewpoint was spread widely in the caliphate by
the end of Marwani times; had own fiqh

Kharijis: rejecting both 'Ali and 'Uthman and looking to a righteous imam
chosen by a righteous remnant of Muslims : taught that great sinners not
merely lacked true faith but were legally infidels ; their doctrines were
spread widely by their guerrilla bands, but the less activist (Ibadiyyah)
centered in Basrah

medina (followed by mecca)

School of Ibn-'Umar: accepted Marwanids for sake of jama'ah, but critically;
often concentrated on collecting hadith reports, emphasized the determination
even of human acts by God ; had beginnings of own fiqh

School of Ibn-'Abbas : inclined to prefer 'Alids to Marwanids (in fiqh were
independent of the moderate Shi'ah of Kufah) ; concentrated on inter-
pretation of Qur'an (strong in Mecca)

SYRIA (DAMASCUS and ALEPPO)

'Adliyyah or Qadariyyah : accepted Marwanids critically; taught the justice of
God and human freedom to do right, in part following Hasan of Basrah ;
especially among the Kalb tribal faction, around Damascus

Jama'is : accepting Marwanids uncritically, were close to the school of Ibn-'Umar
of Medina ; had own fiqh

Same schools as Medina

Jama'is: as in Syria. khurAsAn

Shi'ah : as in Kufah

Jahmiyyah : highly critical of Marwanids ; taught the determination of human

acts by God, and rejected anthropomorphism; perhaps related to

Murji'ah, who in Khurasan had Mawali connections

264 THE CIVILIZATION OF THE HIGH CALIPHATE

human acts, since they form part of God's creation; therefore He alone must
have decreed them, and the human actors can have had no power over their
destiny. Passages in the Qur'an about how God leads further astray those who
have once neglected Him seemed to lend themselves to such a notion. 9 Some
Muslims, especially at Medina, drew the conclusion that human freewill was
illusive. In reaction, others analyzed the meaning of the term for the decrees of
God, qadar, and tried to show that God need not necessarily determine all
human acts; these men were called, by their opponents, Qadaris. (The most
famous man to take this position was Hasan al-Basri, who, as we have seen,
opposed indignantly the deterministic interpretation of the Qur'an that had
become popular in Medina.) 10

The Qadaris began a reform-minded religious party especially among the
Kalb tribal bloc in Syria, which was increasingly alienated from the later
Marwanids. On the test case provided by the fitnah wars, which — as for
Kharijis and Shi'is — became the model in terms of which a party's attitude to
the imamate was formulated, they supported Mu'awiyah against 'Ali, and
hence the principle of the jama'ah; but they reserved the right to criticize the
Marwanid rulers in the name of enlightened religious concern.

In other, related debates, set going especially by the Kharijis, men discussed
the meaning of human submission to God- — and hence the meaning of the
relevant Qur'anic terms, muslim, islam (submitter, submission) and mu'min,
xmdn (faithful, faith): the Kharijis (convinced of the invalidity of rule by bad
Muslims) insisted that islam and imam must carry an observable moral conse-
quence, by which people could judge both ruler and ruled. Men who were im-
pressed with Khariji arguments but were willing to allow professing Muslims,
including rulers, the benefit of a doubt insisted that what counted was the
inner conscience, which only God could judge; they were called Murji'is. (The
most famous of these was AM-Hanifah, the Kufan legist.)" On the problem of

' Though the discussion of Muslims' positions (including an alleged early Islamic dog-
ma of predestination) ascribed to John of Damascus in the Marwani time is clearly apo-
cryphal — see A. Abel, 'Le chapitre CI du Livre des Heresies de Jean Damascene: son
inauthenticite', Studia Islamica, 19 (1963), 5-26 — yet since such problems were mooted
among Christians, and since Muslims and Christians did argue, it seems likely that
such arguments formed a theological stimulus; but they would surely have arisen
independently anyway.

10 Cf. Julian Obermann, 'Political Theology in Early Islam: Hasan al-Basri's Treatise
on Qadar', Journal of the American Oriental Society, 55 (1935), 138-62. There it might
appear that the first to take a position on 'qadar' were men of power justifying their
sins by predestination; this is unlikely.

" The development of Khariji thinking and its interpenetration with other tendencies
has been traced by W. Montgomery Watt in 'Kharijite Thought in the Umayyad
Period', Der Islam, 36 (1961), 215—31, and more summarily in chapters 2 and 4 of his
Islamic Philosophy and Theology (Edinburgh University Press, 1962) — a book which is a
very quick survey dealing chiefly with the early period, and more satisfactory at those
points that he knows personally than elsewhere. Wm. Thomson 'The Character of Early
Islamic sects', Ignace Goldziher Memorial Volume, part I, ed. S. Lowinger and J. Somogyi
(Budapest, 1948), pp. 89-116, brings out the relation between Kharijis and Murji'ah.
In several places, Watt has developed an improbable thesis that Shi'ism can be

THE ISLAMIC OPPOSITION 265

the fitnah wars, in particular, they held that judgment on both 'Ali and
Mu'awiyah was to be suspended; neither should be condemned. Accordingly,
the subsequent Marwani imamate likewise could not be condemned in advance,
though the rulers must be called to account for particular acts. The Murji'is
formed a reform-minded party especially at Kfifah (and Basrah), but the
name was sometimes extended to any who supported the Umayyads in the
name of the jama'ah principle of unity.

But the most far-ranging speculation was carried out among the more radi-
cal Shi'is (those recognizing imamate by nass designation) ; this was apparently
favoured by the development of relatively closed groups of followers of parti-
cular lines of nass imams. Those most active in this speculation were called
retrospectively, by later Shi'is, the Ghuldt, the 'exaggerators', at a time when
many of the positions they had held were regarded as unsophisticated exag-
geration of what came to be regarded as proper Shi'i views. These Ghulat
theorists tried to work out, above all, the religious implications of a historical
situation in which truth and justice seemed to be represented by a defeated
minority, the Shi'ah, overridden by an arrogant majority in the name of Islam
itself.

The notion of the imam, with his special divine 'ilm and his destiny of bring-
ing true Islamic justice to the oppressed, seized the imaginations of many,
especially of the Muslim Mawali. Among the radical Shi'is, not only 'Uthman
and the other Umayyads were regarded as cursed usurpers, but even Abu-
Bakr and 'Umar, who should have yielded to 'All. 'Ali and his family, on the
other hand, became almost superhuman heroes. A messianic role was expected
of several of them — first Ibn-al-Hanafiyyah and then others were, by this or
that faction, thought not to have died but to have departed, awaiting the
moment of return, when they should carry all before them and 'fill this world
with justice as it is now filled with injustice'. The title Mahdi, the 'Guided'
(as used by Mukhtar for Ibn-al-Hanafiyyah in his rebellion against Ibn-al-
Zubayr), at first merely a designation for the correct Islamic ruler, sometimes
took on eschatological tones. Even before the coming of the imam who would
be Mahdi, moreover, the current imam by nass was felt by many to have a
special relation to God — at least, to receive immediate divine guidance rather
as had Muhammad himself. Without rejecting the unity of the Muslim Um-
mah, something of the expectation of continuing prophetic revelation or at
least guidance, such as had been suppressed in the Riddah wars, seems to have
been revived. Only through continuing contact with the original source of
'ilm could the continued purity of the 'ilm be guaranteed. Then such notions
became the starting point for speculations about the status not only of the

traced to 'south' Arab tribes harking back to Yemeni kingship and desiring a charis-
matic ruler, while Kharijism represents 'north' Arab tribes desiring a tribally charismatic
community; whereas actually a present charismatic individual proved more crucial for
Kharijis than for Sht'ls. He seems to have accepted certain misconceptions about the
cultural role of ancestry (as well as about its factuality). (On the Murji'ah, compare also
my article on the Ghassaniyyah in the second edition of the Encyclopaedia of Islam.)

266 THE CIVILIZATION OF THE HIGH CALIPHATE

imam but also of the minoritarian faithful themselves. Already in Mukhtar's
time at Kufah during the second fitnah, recourse had been had to special
divination. Later, with more sophistication, several teachers seem to have
stressed the ability of the ordinary soul, if selected out from the crude world
by its loyalty to the cause, to come angelically into touch with God; and cor-
respondingly, to have stressed a present spiritual, and not merely a future cor-
poreal, resurrection from the dead — in the light of which worldly defeat could
seem less overwhelming, the cause of justice more enduring, and future super-
natural success more promising.

In such perspectives, the role of the developing Shari'ah law, emphasized by
most of the Piety-minded, became less important for some of the Ghulat
theorists (though not for all). What mattered was loyalty to the cause and to
its representative, the imam; preoccupation with legal minutiae was at best
futile in the presence of massive injustice, and at worst might be distracting.
Some taught that the apparent wording of the Qur'an, wherever it dealt with
external minutiae, had a symbolic meaning of a more spiritual import — or
perhaps a hidden reference to the 'Alid loyalist cause itself: behind the zdhir,
'externals', of the revelation which the superficial majority knew and took
literally, lay a bdtin, 'inward meaning', which the majority were blind to and
which only those loyal to the imams could know. In this way, side by side with
the germs of a future Shari'ah-minded interpretation of Islam were sprouting
the germs of an equally potent future inwardly-turned interpretation of Islam,
which was to come to fruition especially within the mystical movement called
Sufism. And just as among the legists and those close to them many elements
were broadly modelled on the traditions of earlier religious communities, so
among some of the Ghulat theorists notions entered in that had been long
current, especially in heretical minority groups. 12

All of the more adventurous of these early thinkers among the Muslims
tended to be under suspicion as disruptive of Muslim unity and introducers of
bid'ah (innovation) ; the more so, as they tended to represent the most active
of the Piety-minded in formulating the general principles of opposition to the
dynasty. Several were executed on the grounds of holding a Qadari position
(that is, maintaining an oppositional mood within Syria itself), others for too
ardent and obviously subversive an advocacy of Ghulat principles about the
Shi'i candidates. (Murji'is and more moderate Shi'is might be oppositional too,
but compared to their rivals, in the Iraq, they were less frightening to the
dynasty.) Later, after the revolutions which overthrew the Marwanis and in
the post-revolutionary reaction, the positions they had held, or at least the
names given to those positions, were widely discredited; Qadaris, Murji'is, and
Ghulat alike, along with most other schools of thought of the time mentioned
on the chart, were supposed to have been heretical (and the more revered

12 Cf. my article on the Ghulat in the second edition of the Encyclopaedia of Islam, in
which will also be found an evaluation of the evidence presented by some (not all) of the
Muslim historians of dogma.

THE ISLAMIC OPPOSITION 267

figures, like Hasan al-Basri and Abu-Hanifah, were gradually dissociated from
such labels). Nevertheless, their thinking formed the point of departure for
Islamic thinking thenceforth. 13

The Marwani dynasty loses its mandate

Mu'awiyah had still been able to represent for most Muslims the unity of
Islam, despite much hostility toward him personally. 'Abd-al-Malik gained
much personal approval at Medina, but he and especially al-Walid faced a
rising hostility from the Piety-minded, above all in the Hijaz. Al-Walid's
governor at Mecca, Khalid al-Qasri, took great pains to honour the holy city
and to maintain a certain religious gravity there, despite the mood of song
and dance which, with the advent of the wealth from the conquest, had made
Mecca the most renowned centre for singing girls and love lyrics. He insisted,
for instance, that men and women must be separated in performing the rite of
running in a circle round the Ka'bah, to prevent inappropriate jostlings. Yet
when he built, at al-Walid's command, an aqueduct to bring abundant sweet
water to the town, the Piety-minded party made it an occasion of complaint.
The brackish well of Zamzam, which had a part in the hajj as the most impor-
tant well in Mecca, had taken on a holy aura for Muslims (as presumably it
had earlier for the pagans) ; it was regarded as the work of Abraham and as the
well which Muhammad had used. Al-Walid's aqueduct would now rival Zam-
zam; it proved how impious was the reigning dynasty.

Al-Walid's successor, his brother Sulayman (715-717), was still more readily
despised by the pious; he was notoriously pleasure-loving. (He also had a
streak of cruelty; for instance, he enjoyed watching captives being hacked to
death with dull swords. Hereafter this sort of taste began to recur among
hereditary rulers.) Perhaps worse for the state, he allowed party spirit to com-
promise his position as representing Arab unity. In the great strife between
the tribal blocs, he allowed his governor of the Iraq and Khurasan to favour
the Qahtan Arab faction, exercising special cruelty against the family of al-
Hajjij, who had been of the Qays faction. Henceforth, in the eastern pro-
vinces, governors found themselves driven almost irresistibly to depend on
one or another tribal bloc for support during their term of office — and in con-
sequence to incur the opposition of the other bloc. Nevertheless, in 716 the

13 I owe much of this discussion of early schools of thought to Wilferd Madelung, who
has supplemented orally his discussions in Der Imam al-Qasim ibn Ibrahim und die
Glaubenslehre der Zaidiien (Berlin, 1965). (But he is not responsible for my inter-
pretations.) It is a rich, sound, thorough book which yields more than its title suggests.
On Jama'is, Murji'is, Qadaris, and Jahmis, see his masterly appendices, indispensable
to an understanding of early Muslim factions. The Murji'is are especially clearly
analyzed; but it seems rather mechanical to suppose, as Madelung does, that any partici-
pation in rebellion on their part was necessarily contrary to their principles. His identi-
fication of the Hadith folk of 'Abbasi times, from al-Rashid on, with the pro-Umayyad
jama'ah party may be oversimplified.

268 THE CIVILIZATION OF THE HIGH CALIPHATE

Arabs were able to force the submission of the Iranian princes in Tabaristan,
the rainy lands at the southern end of the Caspian.

But under Sulayman, the wars with Byzantium led to a point of crisis.
Successful as were Arab arms under 'Abd-al-Malik and al-Walid in Spain (and
even, a bit later, in Gaul up to a point), in Sind, and in the Syr-and-Oxus
basin, the most important enemy was always the Byzantine empire, whose
heartland was the nearby Anatolian highlands. No further attacks were
launched against Abyssinia, relatively distant, poor, and inaccessible; but the
capture of Constantinople remained a primary goal of the Muslims; to have
seized the Sasanian empire and yet let the main parts of the Roman empire
escape seemed only half a victory. As soon as 'Abd-al-Malik found himself free
of internal enemies, he broke off the truce he had made with the Byzantines,
defeated them soundly (692), and launched yearly attacks into Anatolia.
These carried the Arabs steadily further until in 717 (under Sulayman) they
were ready again to besiege Constantinople. The Syrian army and a large part
of the ready resources of the empire were committed to the effort; it failed,
and the Marwani power was seriously weakened for the moment; it was the
last attempt on Constantinople by the caliphal state. 14

Whatever his inadequacies, Sulayman was very devout; he respected great-
ly one of the Piety-minded preachers resident at Damascus, who persuaded
him, in his last illness, to alter the succession established by 'Abd-al-Malik.
After a reign of but three years, therefore, he left the caliphate to a pious and
upright cousin, 'Umar b. Abd-al-'Aziz (717-720). 'Umar II (as he is called) had
been closely associated with the Piety-minded groups at Medina, and tried to
carry out the spirit of their policies. He was able to win the support of all
elements of the incipient Piety-minded party, even Kharijis and some Shi'is;
at the same time, he retained the respect of his own family, the Marwanids
and the house of Umayyah; — he was looked to as a model, in some degree, by
his successor. His brief reign offers a glimpse of the lines along which the
Islamic vision of the time might have been more fully embodied in practice.

He tried to model himself on 'Umar b. al-Khattab, though not blindly (for
sometimes he reversed even what had been acts of 'Umar himself which he
regarded as injustices). 15 In the first place, he scrupulously accorded the old

14 H. A. R. Gibb, 'Arab-Byzantine Relations under the Umayyad Caliphate',
Dumbarton Oaks Papers, no. 12 (1958), pp. 219-33, and reprinted in his Studies on the
Civilization of Islam (Boston, 1962), points out that from this defeat on, the caliphal
policy — economic and administrative — was directed toward things that would make
their state a successor to the Sasanian, rather than to the Byzantine; or, as I would
put it, more toward internal agrarian centralization than toward an expansive Mediter-
ranean foreign policy.

15 Muslim historians have generally given 'Umar II high honours, but for a time
Western historians thought him a pious fool; the evidence for his practical sagacity is
gathered by Julius Wellhausen in the chapter on 'Umar II in Das arabische Reich und
sein Sturz (Berlin, 1902). For a recent confirmation of 'Umar II's combination of piety
and good sense, see H. A. R. Gibb, 'A Fiscal Rescript of 'Umar II", Arabica, 2 (1955),
1-16.

THE ISLAMIC OPPOSITION 269

Medina families the special status 'Umar had granted them; in particular he
admitted their legal claims to certain properties which they had forfeited as
the caliphal state was being consolidated. He ended the condemnation of 'Alt
from the pulpits, which the victorious Mu'awiyah had instituted to reinforce
unity but which now had the reverse effect. Indeed, whoever could show that
his rights by the original settlements of the conquest had been later abridged,
or that he suffered other unjust settlements, could obtain redress. A Berber
tribute in children was abolished; some Christian groups had their tribute
dues reduced; church lands in Egypt were freed of certain taxes; other illegal
taxes in Iran were remitted; some excess taxes already paid were restored. He
managed to find effective governors who yet ruled without brutality to the
persons of the Muslims and who abstained from building up private fortunes.

Perhaps his most important policy was one of quietly treating all provinces
alike. In particular, he removed from the Iraq the more obvious evidences of
Syrian dominion, perhaps even many Syrian troops; he gave some remoter
provinces more local control of their revenues. The central budget was re-
duced (despite a programme of charities which extended to all provinces, not
just chiefly Syria as had previous Marwanids' charities), partly by eliminating
jihad wars on most of the frontiers, which in many cases had become glorified
plundering expeditions without permanent results. (On the Byzantine frontier,
however, a policy of peace was dictated by sheer prudence.) The chief agitator
of tribal rivalry under Sulayman was jailed.

But perhaps 'Umar II 's most heartfelt concern was to encourage general
conversion to Islam. Already under 'Umar there had been a clear desire that
all Arabs be Muslims; and other pastoralist groups, notably the Berbers, had
been assimilated to the Arabs in this respect and had joined them in their
conquering expeditions. Occasional Muslims had encouraged or forced dhim-
mis (non-Arab non-Muslim subjects) to convert, especially in the case of in-
dividuals or families whom it was important to attach to the Arab cause. Now
conversion became a government policy extended to all the non-Arabs. It was
required that village heads in Egypt be chosen from among Muslims. Even
while scrupulous justice was extended to them, within the terms set by the
Arab conquest, Christians were made to feel inferior and to know 'their place'.
It is likely that some of the humiliating sumptuary laws that later were some-
times imposed on the wealthier dhimmi non-Muslims (and Actively ascribed to
the first 'Umar) were sanctioned by 'Umar II: that Christians and Jews should
not ride horses, for instance, but at most mules, or even that they should
wear certain marks of their religion in their costume when among Muslims.
'Umar II has passed in Christian tradition for a persecutor. Though he dis-
paraged jihad for plunder, 'Umar II was eager to persuade frontier princes to
become Muslims, if necessary by the lures of interest. Sometimes because
dhimmis assessed the taxes, and latterly because governors like al-Hajjaj had
been unwilling to admit Muslim converts, especially of the lower class (whom
he wanted to stay on the land as labourers to boost agriculture), many upper-

270 THE CIVILIZATION OF THE HIGH CALIPHATE

class dhimmts had continued to be free of poll tax while many less powerful
Muslims continued to pay it. 'Umar II insisted that all dhimmis and no Mus-
lims pay the poll tax (to which the term jizyah was eventually restricted).

A complaint of the converts and their descendants, the Muslim Mawali, was
that though some of them had fought in the Muslim armies, helping in the
conquests, they had not been included in the army diwan so as to draw a
share of the revenues. In Kufah in the second fitnah, the Shi'i Mukhtar had
bound them to him by so including them — and had enraged the Arab Kufans.
'Umar II did not include the descendants of the Kflfan Mawali, but he did
allow Mawali who had personally participated in the fighting a (limited) share
in the booty from the most recent conquests, for instance those from Khurasan.

In his land tax policy, 'Umar II's concern for conversion was reinforced by
another concern — to avoid the growth of an unduly privileged sector within
the Muslim ruling class. Landowning Muslim Arabs within Arabia paid a land
tax in the form (more or less) of a tenth of the produce — a tithe ('ushr). (This
was in addition to the zakat tax paid by Muslims on property other than land.)
But most land tax derived from the conquered lands, which normally paid
much more than a tenth. The tax on such lands was called khardj, in contrast
to the tithe, and went in principle to the Muslims who had conquered them
and to their heirs, according to the army diwan. It was naturally felt that only
dhimmis, not Muslims, should pay the kharaj — Muslims, who formed the
soldiery on which the caliphal state relied, should be on the receiving end. As
gradually more and more Muslims acquired kharaj-paying lands — and as,
alternatively, kharaj-paying dhimmi landowners became Muslims — what hap-
pened to the taxes varied from place to place and from occasion to occasion.
Al-Hajjaj had gone to an extreme in defending the state revenues by insisting
that all land once kharaj land should pay kharaj whoever owned it; otherwise,
the revenues coming to all the Muslims as a body, and divided among them,
would be gradually reduced, to the profit of those few Muslims who happened
to acquire the formerly kharaj-paying land and need pay only the tithe on it.

Al-Hajjaj's measure had been bitterly resented by the more privileged Mus-
lims and had therefore added to his ill-repute at Medina. But 'Umar II recog-
nized that the problem was real. He rejected al-Hajjaj's measure as reducing
Muslim and dhimmi to one level, but he was equally unwilling to let the
wealthier Muslims monopolize the fruits of the conquest. He allowed lands
already acquired to continue paying only the tithe. But from the principle of
'Umar I, that the Muslims generally should receive the revenues as a lump
sum rather than divide the lands among themselves, he derived a new rule
that Muslims henceforth (after the year 100 of the Hijrah) should cease acquir-
ing kharaj land. As for converts, he evidently provided that in relevant cases
they should give up the kharaj land they personally held — as belonging to all
the Muslims — though they might still work it, paying through their village
the amount of the kharaj, not as kharaj but as rent. Converts to Islam at that
date commonly were not landowners primarily, but went to the new Muslim

THE ISLAMIC OPPOSITION 271

towns to reside; accordingly, this measure was a fairly realistic adaptation to
the time of 'Umar II of the conception of 'Umar I that the Muslim Arabs
should form a separate privileged community, benefiting from taxes paid by
the dhimmi countryside as a mass. Now, however, with non-Arabs freely ad-
mitted, the emphasis was more clearly on the Islamic allegiance as such.

With warfare almost eliminated, there being neither revolts nor frontier
raids, 'Umar II's treasury stood up well under the considerable demands he
made on it by his generous measures. But after less than four years, at the
age of thirty-nine, he died. Some of his measures seem to have been maintained
in principle, but they were not well enforced. Power reverted to Sulayman's
brother Yazid II (720-724), given to women and song. A revolt at Basrah
greeted him, which was not only carried on in the name of Islamic justice, but
was sufficiently influenced by the scruples of pious theorists (referred to as
being Murji'is) to be hampered in its tactical operations. (They insisted on
not assaulting fellow-Muslims without discussion first.) But it was primarily a
movement of the Qahtan tribal faction, and its defeat was followed by a parti-
san rule favouring the opposed Qays faction throughout the eastern provinces.

Under the devout Hisham (724-743), who restored a strong hand to the
helm, relations between the dynasty and its miscellaneous opposition did not
mend. For meanwhile, the underlying development of an integrated agra-
rian-based society, calling for a strong central rule, continued to make for just
the sort of monarchy the Piety-minded objected to. Even the reign of 'Umar
II, with its tendency to equalize Arabs and Muslim Mawali on the basis of a
common Islam, further contributed to the integration of the Arab rulers into
the regional life, despite some of his intentions. By the time of Hisham, and
partly through Hisham's own labours, an impressive amount of bureaucratic
organization was directly in Muslim hands and centrally controlled from the
caliph's capital. Hisham was surrounded by high officials who stood between
him and the commons, Muslim or not. The early 'Abbasid caliphs, masters in
absolutism, subsequently acknowledged him as an administrative model.

The absolute temper of his rule was illustrated in an institution (the musd-
darah) which had been used as early as 'Abd-al-Malik but was more usual and
perhaps more essential in Hisham's time. The governors and heads of financial
bureaus were under temptation to enrich themselves at the expense of the
treasury. It became an established practice that an official who incurred the
special displeasure of the caliph might be discharged and arrested in a moment,
subjected to a more or less arbitrary scrutiny to determine the amount he
might be supposed to have embezzled, and fined a corresponding sum; it be-
came common to use torture either to discover the amounts embezzled or,
especially later, to force the official to disclose the places where or the mer-
chants with whom he had deposited his wealth, so as to collect the fine by con-
fiscating it. Such personal degradation of a high-placed Muslim would have
appalled 'Umar or Mu'awiyah. It was inconsistent not only with the Islamic
standards of the Piety-minded opposition but with the personal ideals of free

272 THE CIVILIZATION OF THE HIGH CALIPHATE

Arab tribesmen. But those who wished no limits to be imposed on monarchi-
cal authority regarded such a threat as a salutary curb on those overly ad-
vantaged in society.

Hisham did not alter 'Umar IPs fiscal principles as such. But in carrying
forward the process of administrative and fiscal consolidation, he proved both
tight-fisted and greedy. His tighter fiscal measures provoked numerous re-
volts, which in some areas (notably among the Berbers) took a Khariji form.

The results of the absolutist tendency were exacerbated by the empire's
being based eccentrically in Syria: unless controlled by a masterly chief like
'Umar II, it came to imply increasing control by that one province over the
others, and by the time of Hisham everywhere it was garrisons of Syrian
troops that counted militarily. The Iraq witnessed the rising of several Khariji
bands protesting public religious laxity, but it was especially alive with
threats of Shi'i revolt centred on Kufah. Khalid al-Qasri, who as governor of
Mecca had been unable to please the zealous, now as governor of the Iraq
faced active anti-Syrian plotting based on the new ideas of 'ilm and imamate;
in 737 he seized two such leaders, with a handful of followers, and burned
them. He was unable, however, to get at perhaps the most dangerous Shi'i
organization, that which placed the imamate in the line of Ibn-al-Hanafiyyah
and by Hisham's time acknowledged the leadership of the 'Abbasid family.
Such organized Shi'ism was not yet predominant, but it seems to have sapped
the strength of more moderate Shi'i movements, notably that of Zayd b. 'All.
A chief of one of the 'Alid families in Medina, in 740 Zayd led in Kufah a futile
insurrection which, however, had wide support elsewhere also.

The 'Abbdsi compromise

Hisham's nephew and successor by Yazid II's arrangement, al-Walid II
(743-744) was not a strong ruler; moreover, he was notoriously careless of
religion. Perhaps worse, like Yazid II he was partisan to the Qays faction and
offended the Kalb tribesmen sufficiently to evoke a rising against him in Syria
itself. He was killed and another Marwanid (Yazid III) made caliph instead
with the support of the Kalb tribal faction and of some of the Piety-minded;
but the untimely death of the new reforming caliph soon followed. His brother,
who took his place, was less able.

Meanwhile, a Marwanid general of a side branch of the family, Marwan b.
Muhammad (744-750, styled Marwan II out of respect to his ancestor Mar-
wan), rose to avenge Walid II in the name of Qays sentiment; for his military
power was based largely on Qays tribal alignments in the northern frontier
areas. He was the most important of the commanders against the Byzantines
and hence the strongest single military chief. But he stood for faction rather
than for Muslim unity. The field was wide open for whatever movement could
seize the initiative, especially if it could be identified with the anti-Qays fac-
tions. Two Khariji movements, one in southern Arabia and the other in the

THE ISLAMIC OPPOSITION 273

upper Mesopotamian plain (the latter, as ever, identified with the Rabi'ah
tribal faction) gained large numbers of supporters, and each for a time seemed
destined to triumph. Marwan II proved able to out-general them. He had
done much to improve the Arab military organization in his frontier wars,
concentrating his men in solid drill units. With his experienced troops, first he
suppressed the latest Syrian Marwanid candidate (and eventually wiped out
the Syrian power altogether) ; then he turned to his rival movements of revolt
and crushed them one by one.

Along with the two Khariji attempts failed also one important Shi'i attempt.
Its imam, 'Abd-AUah Ibn-Mu'awiyah, was a descendant of Abu-Talib, 'All's
father, and hence a member of Muhammad's house of Hashim. (Some are said
to have supposed he would give the rule to an Alid when he had won.) He
made Kufah his first centre of revolt and gained general Qahtan tribal support ;
when hard-pressed there, he retired to the Iranian highlands and staved off
full defeat for some time, gaining the support of representatives from almost
all elements of the Piety-minded opposition. But Marwan's disciplined Qays
troops proved too much for his ill-assorted malcontents.

Meanwhile a second Shi'i revolt was replacing his (also with Qahtan sup-
port). Ibn-Mu'awiyah may have claimed the blessing, in his endeavour, of
Abu-Hashim, the deceased head of the partisans of 'All's son Ibn-al-Hanafiy-
yah; but it seems likely that Abu-Hashim had actually bequeathed his
authority, which carried with it organized support, to another family of the
house of Hashim, the 'Abbasids, descendants of Muhammad's cousin 'Abd-
Allah b. 'Abbas. The 'Abbasids, at any rate, were carrying on a very effective
propaganda against the Marwanids on a largely Shi'i basis.

Tribal-bloc jealousies had been peculiarly strong in Khurasan, the high-
lands northeast of the central Iranian deserts. There the Qahtan faction was
actively discontented. The Mawali formed an important and active element
among the Muslims of that distant frontier province and were jealous of the
privileged position the central Marwani regime still tried to accord the old-
Arab element, despite 'Umar II's decree of equality. A local revolt in the name
of Qur'an and sunnah began in 734, not to be put down till 746 (one of its
leaders, Jahm b. Safwan, was a Piety-minded thinker). The able Marwani
governor, Nasr b. Sayyar, could barely keep order. Into this ferment came
agents of Abu-Hashim's partisans, under 'Abbasi direction, secretly pleading
the cause of the house of Muhammad. The chief of these, Khidash, is said to
have preached radical ideas in the manner of the Ghulat and to have been
disowned by the 'Abbasids; he was executed in 736. But the propaganda con-
tinued till a freed slave of the 'Abbasid imam Ibrahim, Abu-Muslim, was sent
(c. 745) to arouse active revolt among an already substantial group of par-
tisans. 16

18 I use the term 'Abbasid for the Banu 'Abbas, the descendants of 'Abbas, Muham-
mad's uncle; and the more general term 'Abbasi for the adherents of the dynasty and for
their ways. I make a corresponding distinction for later dynasties also.

274 THE CIVILIZATION OF THE HIGH CALIPHATE

Abu-Muslim seems to have stressed revenge for the deaths of various mem-
bers of the house of Muhammad (the Hashimids) at the hands of the Umayyad
house, starting with Husayn at Karbala' and including most recently the rebel
Zayd and his sons; the latter had been pursued into Khurasan and killed.
(Yet Abu-Muslim himself was said to be responsible for the death of the de-
feated Ibn-Mu'awiyah when that rival Hashimid fled to Khurasan.) The black
banners which he raised, as had some earlier rebels, as an eschatological symbol
seem to have been felt to symbolize mourning for these deaths — and indirect-
ly, therefore, for all the injustices that either old-line Arabs or insulted Mawali
felt they had received at the hands of the ruling Arab factions — personified in
the Umayyad house. By 747, he was ready to emerge from secret agitation
and take the field. Even then, canvassing support in the name of the house of
Muhammad generally, he named no candidate to the caliphate, seemingly
leaving that to be settled on victory. Marwan II discovered, however, the
connection of the 'Abbasid Ibrahim with the revolt and had him put out of
the way. Abu-Muslim added him to the list of Hashimid martyrs. With the
defeat of the other anti-Marwani movements, Abu-Muslim's was the only
opposition movement still in the field by 748 and attracted wide support.

Against Abu-Muslim's Khurasanis, Marwan's military reorganization and
good generalship proved not enough. Abu-Muslim proved a talented states-
man, able to mobilize excellent generals. In any case, Marwan II was too busy
suppressing other movements to deal adequately with so distant a danger,
despite his governor's urgent pleas for help. By the time he could turn his full
attention to it, it was overwhelming and swept all before it. Abu-Muslim
moved through western Iran with little opposition; at the Euphrates, a major
Marwani army was defeated. Abu-Muslim's men now brought the 'Abbasids
into the open as claimants, proclaiming Abu-1-' Abbas, Ibrahim's brother, as
caliph (with the byname of al-Saffah) in the heart of Shi'i territory at Kufah.
Muhammad's house of Hashim was to rule at last (749). Marwan's whole
power was defeated near the Tigris (750) ; he was unable to raise new forces
among the Syrians, whom he had himself crushed; at last he was trapped and
killed in Egypt. Gaining an almost unquestioned domination as far as Egypt,
the 'Abbasi movement thus ended the series of revolutions that the Syrians
themselves had begun."

All but one of these movements — all, that is, but Marwan II's — were
coloured by the demands of the Piety-minded for a new and more adequately
Islamic social order. That of the 'Abbasis was keyed to Shi'i expectations.
Their imam had been supposed to have authority by virtue of his connections
with 'Ali by bequest and with Muhammad's house by nature (for his ancestor
'Abd-Allah b. 'Abbas was, like 'Ali, Muhammad's cousin) ; he was said to have
foretold, through the 'ilm knowledge so acquired, many detailed events of the

17 Claude Cahen, 'Points de vue sur la revolution "abbaside" ', Revue historique,
230 (1963), 295-338, has the most careful recent study of the 'Abbasid cause and its
leadership of the Shi'i movement.

THE ISLAMIC OPPOSITION 275

time of the revolutions. But when the 'Abbasids had seized power they turned
their backs on their more strictly Shi'i supporters, for the most part, and made
relatively little of their role as imams endowed with 'ilm. Nor did they make a
serious effort to rule according to the ideal religious attitudes cultivated by
the pious among the opposition. Instead, they based their power on the force
of their Khurasani supporters; and above all on their ability to provide a much
desired peace after all the revolutions, on the basis of acceptance of the chief
practical demands of the main ethnic elements of the opposition to the Mar-
wanis. They showed from the very first that their rule, far from embodying
ideals of egalitarian personal responsibility, was to be one of bloodshed. Al-
most their first care was to massacre all accessible members of the scapegoat
Umayyad house. But these were soon not the only ones to suffer. Whatever
else their rule might do, it would brook no opposition to itself.

The 'Abbasids took the power away from the Syrian Arabs and distributed
it more widely; though they favoured the Khurasanis, they placed their capital
in the Iraq (building for that purpose the city of Baghdad), and made it clear
that no region, unless perhaps Syria itself, was tabe discriminated against. At
the same time, they made no effective distinction between the old Arab
families and the new Muslims, the 'Mawali', who had come up from the con-
quered population (and many of whom no longer even had any affiliation with
an Arab tribe). In this way they satisfied some of the most pressing demands
of the opposition. On a more idealistic level, they were content to offer to the
Piety-minded groups a de jure recognition of the legal programme they were
working out, and were willing to honour and occasionally pay attention to the
representatives of hadith reporting and Islamic 'ilm. In this latter point they
differed from later Marwanids like Hisham, among whom representatives of
the Medina piety were by no means despised. But the 'Abbasids were will-
ing to accord formal and exclusive status to the representatives of the
former Piety-minded opposition. They tried to appoint as qadis, for instance,
men whom the pious would recognize as representative of the new ideas about
the sunnah, and some (but not all) such men were willing to accept the ap-
pointments. At any rate, the qadi courts were to be bound by the legal fiqh of
the Piety-minded. In those several ways, the old historical constellation under
which the Piety-minded opposition had developed was broken up. The support
the Piety-minded groups had derived from ethnic complaints disappeared,
and they were themselves divided. The old Piety-minded opposition could no
longer expect to institute their demands on the basis of the general sentiment
of the Muslims.

The Shi'is, and the Piety-minded opposition generally, were disappointed.
Sensing or anticipating this disappointment, indeed, the 'Abbasids soon did
away with their leading Shi'i supporters, paralyzing immediate resistance to
their policies. The head of the Shi'i group in the Iraq (Abu-Salamah) had
evidently hoped, on the triumph of the 'Abbasi armies, to proclaim one of the
chief 'Alids caliph; he accepted the 'Abbasi chief only with reluctance when

276 THE CIVILIZATION OF THE HIGH CALIPHATE

no 'Alid proved ready or adroit enough. Not surprisingly, ground was soon
found to have him killed. More disquieting was the death of Abu-Muslim him-
self. He was on bad terms personally with al-Mansur, who became caliph in
754, soon after the victory. He distrusted the new caliph but was lured by
promises of reconciliation into the caliph's presence alone without his devoted
soldiers, and was murdered; the unprepared soldiers were made to accept the
fait accompli.

The 'Alid loyalists further from the seat of the new government tried to
pull themselves together for a new effort. One of the most prominent of the
'Alid candidates, Muhammad 'al-Nafs al-Zakiyyah' ('the pure soul'), staged a
revolt in the Hijaz. He claimed the title Mahdi, '(divinely) guided' — the
title implying a religious mission, already used by Mukhtar for his candidate
Ibn-al-Hanafiyyah — and he seems to have had the sympathy of Piety-minded
groups everywhere apart from the Kharijis and the particular band of Shi'is
that personally supported the 'Abbasids; even Basrah revolted and was
bloodily reduced. But this revolt, which in the Hijaz, at least, was so pietistic
that it tried to imitate, against 'Abbasi might, techniques Muhammad had
used against Arab tribesmen, finally failed miserably.

The Piety-minded groups, accordingly, were presented with two alter-
natives: to continue their overt opposition but on some new historical basis;
or to accept the settlement offered by the 'Abbasids as a half-loaf. Gradually
the majority accepted the 'Abbasi settlement and adopted the black robes
which, like the black flags, were the emblem of the dynasty. In effect, they
clung to the jama' ah, recognizing the validity of the general community ex-
perience however imperfect the community might sometimes be. Those Piety-
minded who had made a point of not breaking with the majority of the Muslim
community had long referred to themselves as adherents of the jama'ah, the
community as a body. But till now, the various Piety-minded factions had
been relatively local, and their rejection or acceptance of the ruling dynasty
had varied in many degrees. There had been no overall schism in the com-
munity. Now the 'Abbasid house gradually won a position, among most of
the heirs of the Piety-minded groups everywhere, as sole legitimate rulers,
whatever their faults; while those who anywhere still remained unreconciled
found themselves frankly in a minority position. The Muslims' allegiance was
to be polarized throughout the empire, pro or con.

The two sides in this schism have subsequently been labelled, convention-
ally, 'Sunnis' and 'Shi'is'. Each side adopted a historical position in justifica-
tion of its present attitude. The Shi'is came to reject the religious authority of
those, even among Muhammad's associates, who had not recognized 'All's sole
right to the caliphate (a position hitherto held only by radical Shi'is). Corres-
pondingly, it gradually became characteristic for their opponents to accept
the religious authority of all Muhammad's associates without distinction, des-
pite the quarrels that had taken place among them.

The term Sunni is short for 'Men of the Sunnah and the Jama'ah'. This name

o
m

a

•iH

H

u

W

Cd

XI

a

en

«rt

P

«j

M

'2

o

Xi

T3

W

ID -a

-=> 2

•a

X3

W)

ni

XI

a

+->

<D

■a

5

-d

fl

a

rt

<aS

N

a

<(fl

p

»

3 si J!

-<3
("■-J -u

XJ
ft

i-i ci JJ

^£ a

<gS8

S H - o

— a o S

S43

"- 1 JTl "*

ftw

3j=3 (3 ft

J>3§ £

IS

S £

t-X)

C &

xf.S *xi

a c

<3 «nl ti

ft

COS

X2 >'

"^ O

xi bo

!*

»+> N

■SX!

5, wis

t< 0) rt

«

.x

10*0 13
* C C

Xi a

*3,

■3

'G

t-<

Si

M

•— t

<-^

<u

S-a

<ctS

+>

f.

.£

^3

in

^3

CD

a*

03

r-

o

in

4)

<3^

XI ..

ID

r-xi

T3
C3
ca

■ +j

«

2 « < ^3 -S

w I -^ ,%-f.

ft

3
CJ

° m

J!5

<^3
- d

<oi T3 . «

'^3
IS-. m

n 4) ft

M St! a

rv,5.y in

xi g

a «

13 =3 a

Xi -<cj

M 8 ?

!* a>

s.al

^.^3

?> a
as |

: c ■« 5

a £

<^ .S ^3
<D aj

r* +> v

> 4)

1) qj

O

O w

^ ©

©T3

a

£3

MM

to

a

■■""'

CD

^

O

nJ

Ih

w

O

+j

s

HH

ft

e

in

bo

XI +j

u

O

Tl

XI
t— I

1

o

Xi
bo

d

>

•d

ftxi

£

a

B

+j

ri

1)

a

«

3

CO

o

-*-»

a

d 3

•OXI

P

278 THE CIVILIZATION OF THE HIGH CALIPHATE

was first adopted by only one faction among those who accepted the 'Ab-
basids — a faction which stressed continuity with the Marwani past (and was
not especially friendly to the 'Abbasids as such) and combined this with a
special interest in the sunnah practice as expressed in hadith reports about
the Prophet. But since that faction eventually was specially recognized as
representing the jama' ah position, the term has come to refer not necessarily
to all that faction's complex of teachings, but simply to the acceptance of the
jama/ah principle in contrast to the 'Men of the Shi'ah', the 'Alid-loyalist
party.

We do need a term for those who rejected the Shi'i (and Khariji) positions
in favour of the continuing jama'ah; but for this, the term Sunni is inappro-
priate. At best, the term Sunni is confusing, for it has been used, from the
beginning, in special ways by those who wanted to use it exclusively for their
own brand of orthodoxy. Some used it for those devoted purely to the use of
hadith reports {sunnah), without speculative discussion (kaldm). It was used
later, among those who were willing to accept kalam discussion at all, for the
Ash'ari or Maturidi schools of kalam as against the Mu'tazili; it was used by
Shari'ah-minded zealots to distinguish Shari'ah-minded people from the Sufi
mystics; and generally as the equivalent of English 'orthodox'. (A special
disadvantage of the term Sunni is that, unfortunately, laymen commonly
make the mistake of supposing that the Shi'is, because they are contrasted to
Sunnis, do not accept hadith and sunnah.) A far more accurate term would
have been Jamd'i, for the point at issue was the acceptance of the historical
jama'ah unity, whereas all parties accepted the sunnah practice in relatively
similar forms. However, the term Sunni has become so well established in the
sense of accepting the jama'ah, among Muslims and non-Muslims, that one
can hardly displace it at this point. In this work it is not used in the various
more specific ways, but only in the minimal sense, as contrasted to Shi'i. To
stress this usage, I shall use by preference (though rather unhappily) the
hyphenated phrase Jamd'i-Sunni, except in contexts where no confusion can
arise. 18

The Jama'i-Sunni 'ulama' scholars, in this general sense, did not all adopt a
common doctrine; they varied even as to the degree of respect they accorded
the 'Abbasi regime; but they agreed sufficiently under 'Abbasi leadership to
establish eventually a common modus vivendi, and to this day the great
majority of Muslims accept a Jama'i-Sunni position. Those of the 'Alid
loyalists who refused became the Shi'is in the modern sectarian sense, grad-
ually distributing themselves into a number of oppositional sects which have
also persisted to the present. (Many persons more or less loyal to the 'Alids

" We may summarize three ways in which the term Sunni has been most used, as
follows: to mean Jamd'i as vs. Shi'i; to mean Haditht as vs. Kaldmi (including Mu'tazilis
and Ash'aris); to mean Shar'i as vs. Sufi. Then it has been extended to those *Alid-
loyalists, kalam men, and Sufis who accepted key positions of their respective opponents.
Once one no longer assumes the old stereotypes which these usages embodied, they serve
merely to confuse the issues.

THE ISLAMIC OPPOSITION 279

accepted nonetheless a Jama'i position.) The Khariji groups — tending to splin-
ter into mutually hostile bands — likewise continued their opposition, of course,
but it became steadily less effective as the bulk of the population turned
Muslim. 19

19 Leonard Binder and L. Carl Brown have been kind enough to help me clarify a draft
of these early chapters.

The Absolutism in Flower.
7^0-813

From the viewpoint of the Piety-minded, the 'Abbasi regime represented at
best a compromise with their pious ideals for Muslim society — and some as-
pects of 'Abbasi rule, notably its arbitrariness, presented an extreme corrup-
tion of, or even a rude and alien intrusion into, the proper Islamic social order.
The Piety-minded 'ulama' scholars proceeded to develop, in the form of
Shari'ah law and of Shari'ah-minded disciplines harking back to Muhammad
and to the Irano-Semitic monotheistic tradition generally, a programme of
Islamic culture which allowed the 'Abbasi caliphate at best a secondary role.
But from a viewpoint far more popular in court circles and among a great
many of the ordinary population, both non-Muslims and also, now, Muslims,
the 'Abbasi regime represented a reasonably close approximation to a social
ideal. This alternative set of norms cultivated in courtly circles stemmed back
likewise to the early Muslim community, as it had developed under the caliphs
as conquerors, and to the older Irano-Semitic traditions of culture, especially
those of the Sasanian empire. Just as the Piety-minded 'ulama' were develop-
ing a comprehensive cultural pattern, so also did the society surrounding the
caliphal court develop a comprehensive cultural pattern, in which the inci-
pient culture of the Piety-minded could have, at best, only marginal rele-
vance. This pattern — in contrast to that associated with the Shari'ah — was
more aristocratic than populistic; it was based in large measure on agrarian
traditions such as those which had been kept alive from Sasanian times
among the landed gentry of the Iranian highlands, including Khurasan.

The caliphal absolutism

Seen from within this tradition, the caliph was to be a major figure, successor
to the Great King of the Iranian empire close to whose capital Baghdad was
built. He even ought to have a certain religious aura, foreign to the spirit of
the Shari'ah as envisaged by the Piety-minded, but close to that of the old
Sasanians. When the caliph was addressed — as he was — as 'the shadow of God
on earth', the 'ulama' scholars could only be profoundly shocked. The Sasan-
ian monarch, standing at the summit of the divinely ordained aristocratic
society of the Mazdean tradition, had been held to be a special instrument of
the divine will. He had been invested with the sacred divine glory, a mystic

280

THE ABSOLUTISM IN FLOWER 281

aura which represented the authority and power of God. Shar'i Islam, with
its egalitarian insistence that all men were on the same level before God, could
ill tolerate such a figure. Yet the courtly circles were willing to ascribe a very
similar position to the caliph, only limiting themselves to language which did
not go so far as to ascribe to him any part of actual divinity.

In our day, when representative democracy is regarded as the only proper
principle of national government, the monarchical ideal is easily misunder-
stood. Too readily we speak with a certain scorn of 'Oriental despotism'. We
are sometimes surprised to find that most wise men, in both Christendom and
Islamdom, in all ages down to recent centuries regarded monarchy as un-
questionably the most excellent form of government.

In fact, before the advent of Modern technical conditions, a strong mon-
archy was by and large the most satisfactory form of supra-local government
in any agrarianistic society. In very small states — where all that was involved
was a city and its immediate environs, and most people could meet face to
face — in many parts of the world there have existed a variety of viable forms
of municipal governments: rule by one man or by committee, by oligarchic or
by popular assemblies. But once a state gained a certain size, except in those
rare cases where an enduring free federation of municipalities could be created,
monarchy seemed the only suitable alternative to a rapacious armed oli-
garchy. Monarchy became everywhere the acknowledged political ideal;
hence even when, as often happened, the central monarchic authority was
actually too weak to fulfill its functions, nevertheless the forms of monarchy
were retained as window-dressing by the oligarchy which shared rule in its
own narrow interest. '

The principle of monarchy was to give one man in the community the dis-
interestedness of unchallenged supremacy — to make him so highly privileged

1 The most recent attempt to revive the concept of 'Oriental despotism', Karl A.
Wittfogel, Oriental Despotism: A Comparative Study of Total Power (Yale University-
Press, 1957), i s a brilliant and monumental anti-Soviet tract, but so over-schematized
as to be very little use for making intelligible the political institutions of the Arid Zone
(or probably anywhere else.) The previously cited study by Robert Adams, Land
behind Baghdad (University of Chicago Press, 1965), shows how rare, and how limited
as a political model, was the 'hydraulic' type of government based on irrigation to
which a large bureaucratic administration was necessary. In the mid-Arid Zone, some
such pattern was fully effective, probably, only under Nushirvan and his immediate
successors.

Even Wittfogel notes that if any such entity as 'Oriental despotism' can be distin-
guished at all, it is to be found in many other places than the traditional 'Orient' and is
not universal within that 'Orient' ; that is, even he sees that he is discussing not a culture
trait of a particular segment of the world but a phenomenon which may arise anywhere
under the right conditions. But closer examination will carry one further still. Distinc-
tions between absolutisms in Western Europe and absolutisms in other parts of the world
may possibly be of a certain moral importance; but they are of no greater degree, in re-
gard to manner of political functioning, than distinctions between absolutisms in, say,
China and absolutisms elsewhere. To divide all absolutisms into only two sorts, speaking
of 'Oriental despotism' as contrasted (at least in pre-Modern times) to 'Occidental mon-
archy', is to make a false dichotomy.

282 THE CIVILIZATION OF THE HIGH CALIPHATE

that no one could hope to be his rival, so that his interests were no longer pit-
ted against those of other individuals but became merged with those of the
community as a whole, of whose general prosperity he was invariably the
beneficiary whatever the fate of individuals. His authority must be absolute,
one before which the rich and the well-born were as vulnerable as the little
man. He must be able to settle, without hope of appeal, the dangerous quar-
rels of the great; and must have no reason to turn a deaf ear to those who would
plead the case of the poor. Chinese and Arabs, Indians and Occidentals alike
delighted to tell of the petitioner's drum, or bell, outside an olden king's
palace, at the sound of which the king immediately emerged to do justice.
Actuality, of course, was less romantic, but not wholly beside the mark. Such
a monarch might amount merely to a court of final arbitrament; but the mon-
archy would be the more effective, the more the whole administrative activity
of the state could be centred in the hands of his immediate subordinates, his
creatures, dependent on him for all their power. Monarchy was perfected by
bureaucracy. Ideally, such a power could be looked to as the ultimate guar-
antee of equality and justice for the ordinary subjects — the sole recourse for
counterbalancing the natural tendency in society to inequality and privilege.

Long before Sasanian times, at least since the times of the Achaemenid
dynasty of Darius and Xerxes, the Semitic and Iranian peoples had been
evolving their tradition of the absolute monarchy, which became a basic pillar
of the social order of the Sasanian empire. The Sasanian society had main-
tained itself more than four centuries in relative prosperity and with relative
human dignity under a single line of kings. Those who cherished the monarch-
ical ideal looked back to that society as the embodiment of social order and
stability and even of justice for the individual. Its principles were taken as
political axioms. First, the monarchy must be universal — at least it must em-
brace all the civilized lands in its part of the world, in which a rival power might
spring up; for only with such universality could peace be assured among the
various cities and peoples. Second, the monarch must be personally unassail-
able, exempt from anyone's admonition or criticism (lese majeste) ; bound by
the ancestral laws, to be sure, but otherwise by the opinions of no one subject
or clique of subjects. Only so could he wield his mercy and his wrath disinteres-
tedly upon all those who, in his presence, were equally as nothing. Finally, the
monarch must be surrounded by an aristocratic professional staff — heading
the bureaucracy — who (though themselves subject to the arbitrary will of
their master) in turn also were in sufficiently exalted position in society at
large to be able to govern in a relatively detached spirit of noblesse oblige.
The ideal Sasanian monarch was seen in the figure of Nushirvan, who person-
ally embodied both justice and a graciousness toward the humble which marked
off all the more his exalted status. And Nushirvan was seconded by that
wisest of ministers, aristocratic head of the bureaucracy, Buzurgmihr, who
became the model of wisdom for all grand viziers.

But if this kingly personal position was to be assured, as well as that of the

THE ABSOLUTISM IN FLOWER 283

great courtiers, it must be reinforced by a pattern of behaviour which set the
monarch off from all his subjects across a great psychological gulf, as well as
the royal court from ordinary life. Everyone knew well enough that in fact
the king was a mere man among others. In himself he was a mere six feet of
flesh, with passions like any other man, by no means unassailable; and like-
wise his courtiers. Accordingly, the Sasanian monarch had been shielded from
the ready access of his fellow creatures and surrounded with the pomp of
majesty. Now the same was done with the caliph, who from a simple com-
mander among equal believers was raised to a magnificent figure, remote in a
world of awesome luxury, walled off by an elaborate courtly etiquette, whose
casual word was obeyed like divine law. The court etiquette of Baghdad was
consciously modelled on that of the Sasanians and the social implications of it
were essentially the same.

Only the most privileged could normally speak with the 'Abbasid caliph at
all. He could be approached only through a chain of officials and in accordance
with an elaborately formal ritual, which included kissing the ground before
him. It was especially the etiquette of obvious personal submission before the
caliph that roused the ire of the 'ulama' and of the pious generally. Only God
should receive anything smacking of worship. The caliph, being a man, should
be addressed in just the same simple manner as that which the Shari'ah pre-
scribed for anyone else. As hadith reports were found to prove, the Prophet
himself had been so addressed, and who was a mere caliph?

Finally, as symbol of his power, there stood beside him the executioner,
ready to kill the most exalted personage at a word. From the point of view of
the Shari'ah, with its insistence .on personal dignity and a carefully safe-
guarded trial according to God's rules, the caliph's summary executions were
an abomination. From the point of view of the absolutists, they were an essen-
tial means of cutting the Gordian knot of privilege. As with the musadarah
(the fining of dismissed officials and their torture to force payment), the
misuse of summary executions commonly hit only those who voluntarily
frequented the court, had enjoyed its luxury, and had at the same time
deliberately risked its dangers. The ultimate sanction afforded by such
executions was thought to guarantee an effective peace to the wider public.

Such a temper was obviously inconsistent with the personal ideals of the
free Arab tribesmen as much as with the hopes of the Piety-minded. Earlier
manifestations of it had contributed to undermining the support of the privi-
leged Arabs for the Marwani dynasty even while the new converts and their
descendants hated the dynasty for its Arabism. Yet even in the midst of the
third fitnah, Marwan II, in such parts of the empire as he controlled, had
tightened still further the bureaucratic organization and with it the caliphal
absolutism. When the 'Abbasi cause came out victorious, the power of privi-
leged Arabism had already been greatly weakened. Pious ideals notwith-
standing, the way had become open for a still more forthright reconstitution
of the state in terms of the long-standing absolutist civic ideals of the region.

284 THE CIVILIZATION OF THE HIGH CALIPHATE

The first 'Abbasids were completing the work which had already been carried
some distance by 'Abd-al-Malik and Hisham. In al-Mansur, the whole society
could recognize its direct ruler; the Arab families formed but one privileged
element among the rest, the caliphs being as far exalted above them as above
any others.

The foundations of al-Mansur' s power

The first 'Abbasi caliph, al-Saffah, who happened to be the head of the
'Abbasid family at the moment when the 'Abbasi movement seized power in
750, set the 'Abbasi pattern chiefly in the sense that he slaughtered indis-
criminately, treacherously, and, according to the accounts, with gross bruta-
lity as many members of the Umayyad family as he could lay hands on. He
ordered that even the dead be desecrated. Among other tales is this: at one
point he pretended to relent in his bloody search for Umayyad scions and
invited all that remained to a banquet in token of forgiveness. Sitting at the
meal, they were cut down by attendants; a carpet was spread over their dead
and dying bodies and the banquet continued in the same, room to the sound
of their groans. The story is scarcely credible but illustrates what people
thought of the 'Abbasid dynasty.

To the end, the Syrian Umayyads had been careful of their personal and
tribal relations with the other great Arab families; there had often been feuds
and murders and executions, such as that of the Kufan recusant, Hujr; but
everyone felt and behaved as if the killing of any of the more prominent Arabs
was not to be taken lightly. Al-Saffah, on the contrary, was sweeping in his
violence. His power rested less on either Iraqi or Syrian garrison town Arabs,
with their tribal jealousies and personal dignities, than on the Khurasani mixed
Arab and Persian gentry and their peasant troops, many of whom presumably
wanted him to play the absolute ruler as had the Iranian kings of old. His
unconstrained use of power announced unmistakably that his regime would
meet such requirements.

Al-Saffah died (754) a few years after his accession. His brother, al-Mansur
(754-775). succeeded him as head of the 'Abbasid family and hence as caliph.
Al-Mansur proceeded to round out the 'Abbasi absolutist imperial structure.
When by murder he rid himself of the most prominent Shi'is, he was by
the same act ridding himself of men who had figured prominently in bringing
his family to power and therefore were in a position of relative independence
toward it, quite apart from ideals. When he put down the Shi'i rebellion of
al-Nafs al-Zakiyyah at Mecca and forced the 'ulama' to choose either hopeless
opposition or accommodation, he was at the same time asserting the freedom
of the monarchy from limitations on the part of any subordinate group — that
is, of any sector of the privileged elements in the population.

Even so, he was unable to restore so full an absolutism as had prevailed
among the Sasanians. A gifted administrator of his, Ibn-al-Muqaffa' (a convert

THE ABSOLUTISM IN FLOWER 285

from Mazdeism), presented him with a programme that would have restored a
fuller measure of their agrarian-based absolutism. Ibn-al-Muqaffa' urged him
to supplement the direct military basis of his power by rallying the agrarian
classes — in particular, conciliating those families that were still oriented to the
Marwani regime and its values; and by tying the religious specialists to the
state — making the Piety-minded 'ulama' scholars into an officially established
order parallel to the old Mazdean priesthood, and capping such a priestly
structure by asserting a final caliphal authority in questions of fiqh law. Such

The Early 'Abbasid Caliphs

Al-'Abbas ibn 'Abd al-Muttalib, uncle of the Prophet

V

1

I

Muhammad (d. 743)

Ibrahim

Abii-1-' Abbas

('the Imam')

al-saffAh

(d. 749)

(750-754)

Abu Ja'far al-mansur (754-775)

Muhammad al-mahd} (775-785)

Muhammad al-hAd!
(785-786)

Hariin al-rashid (786-809)

Muhammad al-amin
(809-813)

'Abd Allah al-ma'mOn
(813-833)

I
Abu Ishaq
al-mu'tasim

(833-842)

Later 'Abbasid Caliphs

a course was made difficult, however, by the very conditions of 'Abbasi vic-
tory: its dependence on the Khurasani animus against the Syrians, and its
betrayal of the 'Alid cause, popular among many of the Piety-minded. These
conditions, in turn, were the effect of Arab tribal prejudices and the indepen-
dence of the 'ulama.'; it was as champions of individual dignities and foes of
central power that the 'Alids were desired, so that even an 'Alid could pro-
bably have done no more than al-Mansur toward effectively stable absolutism.
The power of Arab prejudices and pious individualism to play so key a role,
finally, was surely, in part, the effect of weak solidarity among the agrarian
classes. Eventually Ibn-al-Muqaffa' was executed on suspicion of heresy. But
al-Mansur did his best.

286 THE CIVILIZATION OF THE HIGH CALIPHATE

Al-Mansur was concerned to control his newly- won empire through a bureau-
cracy capable of minute supervision over every province. He found an un-
scrupulous but highly talented secretary (kdtib, scribe) to organize the whole
financial structure; his work was sufficiently specialized that a single head of
finances such as he was seemed thenceforth usually indispensable, and the
position eventually evolved into that of the all-powerful minister, the vizier.
(Al-Mansur was noted for watching the finances closely — he was ridiculed as
'the penny pincher'; at his death he left a full treasury.) He also was careful
to build up a network of spies to keep down any future conspiracies such as
had raised his own family to power, as well as to control by their information
the various officials with their widely ramifying responsibilities.

All this governmental apparatus was maintained financially from a care-
fully calculated revenue, primarily agricultural. As under the Marwanids, the
agricultural resources of the Fertile Crescent played a major role; but not in
the same proportions. The inner hinterland of the Syrian cities, an arid belt
east of the Orontes and the Jordan, had been highly developed under Chris-
tian rule as part of an east Mediterranean nexus of commerce (and pilgrimage) ;
under the Marwanids it continued to be carefully cultivated, often directly in
the form of dynastic holdings. But already under the Marwanids, the Jazirah,
which as a frontier area had not received much large-scale agricultural invest-
ment, was being developed more fully; and likewise the Sawad of the Iraq,
formerly the basis of Nushirvan's strength. After a time of neglect under the
Medina caliphs, when many of the canals remained almost in disuse after the
disruptions of the last Sasanian years, Marwani efforts, such as those of al-
Hajjaj, seem to have restored the Sawad to full productivity. But under the
'Abbasids, the Sawad came to play almost as central a role in state finances as
under the later Sasanians. The Syrian inner hinterland, disproportionately
expensive to irrigate (and probably at the point of diminishing returns when
such other areas as the Jazirah were opened up) was largely abandoned upon
the ruin of the Umayyad family. Yet the Sawad was not used as a political
base with the same single-mindedness as it had been, evidently, under Nushir-
van. Its direct value to the government seems to have begun declining, in fact,
soon after the 'Abbasid accession, as ever more individuals, intercepted the
revenues. (But this very lack of intensive state exploitation may account for
its having been maintained — despite alternative resources in the Jazirah —
economically at something near the top Sasanian levels and for a longer time:
till the mid-ninth century, when investment in the Sawad seems to have
begun to decrease decisively.) 2

2 Cf. Adams, Land behind Baghdad; and the work of Jean Sauvaget on the Syrian
back country, e.g., L'enceinle primitive d'Alep (Beirut, 1929). There is an unhappy
tendency for global comparisons to be made of economic conditions under the Romans
and under the Arabs on the basis of spotty data. Starting, in fact, from the relative
desolation of the start of the nineteenth century, when Ottoman realms, occupied by
Arabs and ruled by Turks, seemed to show the disastrous consequences of Islam,
Western observers used to note that in 'antiquity' (a period of over a thousand years,

THE ABSOLUTISM IN FLOWER 287

At the same time, the role of merchants and even of craft production was
not irrelevant to the state finances. Merchants paid zakat 'alms' tax (in prin-
ciple, only once a year) on their goods as they transported them through
government checkpoints; perhaps at least as important, they maintained a
network of consumer channels and credit facilities which made possible central
control of monetary taxation on an imperial scale stretching from the Maghrib
and the Yemen to the Oxus basin. Considerations of commercial accessibility,
as well as imperial precedent, seem to have determined the site of the new
capital which al-Mansur proceeded to build, Baghdad, safely away from Shi'i
Kufah and appropriately near the site of the Sasanian capital, Ctesiphon.

Muslims had built cities before. The garrison towns the Arabs had often
built at the first conquests soon became full-scale cities; these were organized
in terms of the tribal contingents that had first settled there. Baghdad was
built on a different plan (anticipated only in part by a new city Hisham had
built) . Its location was not (as in the case of towns like Basrah and Kufah) on
the edge of the desert, of easy military access for a camel-borne people. Like
Ctesiphon, Baghdad was built on the Tigris at a location carefully chosen as
an economic centre for the whole population of the region. It was chosen
partly to command the agriculture of the Sawad, and partly as a crossroads of
bulk and luxury trade; trade by water — along the Tigris and Euphrates rivers
from the Persian Gulf and the east Mediterranean Sea — as well as by land
from across the Iranian mountains. Moreover, Baghdad was not built as the
sum of several independent and equal quarters housing various Arab tribes,
but was focused on the enormous caliphal palace. Its most important part was
the 'round city', the administrative complex in which the caliph lived and
around him the other members of the 'Abbasid family and the various cour-
tiers, each in vast establishments which sheltered large numbers of dependents.
At the outskirts sprang up the bazaars and housing for the rest of the servicing
population. The site was so well chosen that though later caliphs tried to es-
cape its popular pressures as had al-Mansur those of Kufah, it remained un-
rivalled as a cultural and economic centre throughout the High Caliphal
Period.

The first 'Abbasid caliph had adopted a surname of eschatological over-

within which they tended to make no distinctions) areas now desert had flourished
grandly, as the ruins of waterworks and of cities proved. In this context, the scholarly
question arose: at what point did the presence of the Arabs commence its supposedly
ruinous work; and it was a major discovery when it was shown that, at least in Syria,
the land most open to investigation, this was not immediately upon the Arab conquest.
But we are still in want of sufficiently detailed studies according to place and time.
When such are in hand, it will be more easily remembered that the ruin of one area
need not indicate a general decay of the whole region from Nile to Oxus; not only in social
and historical terms broadly, but even specifically in economic terms, a decreasing invest-
ment level in one area may be more than compensated for by an increasing investment
elsewhere. The undeniable economic decline which the region as a whole suffered at some
point between the eighth and the eighteenth centuries must be assessed with subtler
techniques.

288 THE CIVILIZATION OF THE HIGH CALIPHATE

The Flowering of the High Caliphate, 750-813

75° _ 775 The caliphate of the brothers of the imam Ibrahim, al-Saffah

and al-Mansur

THE GENERATION OF THE THIRD FITNAH

7 2 3 _ 759 Ibn-al-Muqaffa' launches adab prose with translations from

Pahlavi
765 Death of Ja'far al-Sadiq, imam of those Shi'is who held to a

Fa/timid line; from his death date their divisions
767 Death of Ibn-Ishaq, biographer of Muhammad

700-767 Abu-Hanifah, great imam of the Iraqi school of fiqh, refuses to

serve under 'Abbasids

786 Death of al-Khalil, first systematic grammarian, prosodist, and

lexicographer

756-929 Spain independent under an Umayyad dynasty of amirs

(929-1031; they are styled caliphs); rivalry of various Arab
tribes and of locally converted Muslims maintains turmoil there

762-763 Foundation of Baghdad, which becomes commercial and

cultural capital of all Muslim territories

775 - 785 Caliphate of al-Mahdi; establishment, gradually, of pattern of

'Abbasi relation to Sunni 'ulama' — acknowledgement of their
type of piety; henceforth, the former Piety-minded Opposition
either come to terms with the 'Abbasis, as Jama'i Sunnis, or
(a minority) eventually go into oppositional Shi'i sects;
persecution of Manicheans, especially as attracting courtiers

THE GENERATION OF THE FIRST 'ABBASIDS

7 1 5-795 Malik b. Anas, imam of the fiqh of the Hijaz

793 Death of Slbawayhi, Basran systematizer of Arabic grammar

798 Death of Abu-Yusuf, major successor in fiqh of Abu-Hanifah,

along with Muhammad al-Shaybani (d. 805)
801 Death of Rabi'ah al-'Adawiyyah, of Basrah, ex-slave-woman

ecstatic in God's love
786-809 Caliphate of Harun al-Rashid; Sasanian court tradition brought

to its height under Barmakid viziers (fall of Barmakids al-Fadl

and Ja'far, 803)

THE GENERATION OF AL-RASH!d'S TIME

813 Death of Ma'ruf al-Karkhi who brought Sufism to Baghdad

816 (?) Death of Abu-Nuwas, dissolute representative of the 'new'

Arabic poetry in contrast to the pre-Islamic and Marwani type
767-820 Al-Shafi'i, at Baghdad and in Egypt, consolidates doctrine of

Muhammad's legal authority and founds, as imam, a personal

school of fiqh distinguished from those of the Iraq, the Hijaz,

and Syria
823 Death of al-Waqidt, pioneer historian of the early Muslims, a

major inspiration of the historian al-Tabari
c. 828 Death of Abu-l-'Atahiyah, poet of philosophic renunciation

809-813 On death of al-Rashid, partition of empire between his two sons;

civil war; the empire is forcibly reunited when al-Ma'mun, with

Khurasan! troop support, defeats al-Amin at Baghdad

THE ABSOLUTISM IN FLOWER 289

tones, al-Saffah, which in itself implied at once nobility and bloody ruthless-
ness in exacting divine vengeance. The name al-Mansur was likewise such a
surname, implying that the caliph was singled out for divine help in his vic-
tories. The Marwani rulers to the end had been known by their simple given
names; al-Mansur made exalted surnames customary in his line by giving one
officially to the son whom he made his heir. He called him al-Mahdi, a title
which the Shi'is had used for the expected restorer of Islamic justice. In this
way he may have implied that the son would make up for the bloody ways in
which the father had established his power, but certainly in effect gave notice
that in any case 'Abbasi absolutism was to be the definitive outcome of the
hopes and plans of the Piety-minded.

The Manichean temptation

Al-Mansur passed on a relatively subdued and peaceful empire to al-Mahdi
(775-785). He had failed to take over only Spain, where an escaping Umayyad
set up a somewhat precarious independent amirate, and the western parts of
the Maghrib, where Hisham had not succeeded in subduing Berber rebellions.
Al-Mahdi made no great attempt to recover those distant provinces. In the
main he continued his father's policies but with less than the paternal parsi-
mony.

His reign furthered the rapprochement between the court and the dis-
appointed Piety-minded factions through both words and deeds, though in a
way which contrasted sharply with that of 'Umar II. The proclamations of
the Marwanids, justifying their own rule in the name of Muslim unity, had
commonly breathed a positive self-confidence in the active political destinies
of Islam as it stood; 'Umar II also had expressed this confidence in his work.
The discourses ascribed to al-Mahdi seem to reflect, more than those even of
al-Mansur, a new and less buoyant attitude of the Jama'i-Sunni 'ulama': that
the days of the first caliphs had been a time apart which the present could not
hope to rival. And al-Mahdi made no attempt to soften a growing absolutism
that seemed unavoidable.

Al-Mahdi's piety went beyond words, however. It was expressed in a rigor-
ous communal spirit, championing the exclusive claims to truth of the true
community. He did not fail to push (without great result) the raids on the
frontiers against the Byzantine empire; but he waged perhaps as vigorous a
campaign internally in a religious persecution of the Manicheans. These pre-
sented, in fact, an attractive alternative both to the Muslim communal spirit
and to the sort of piety represented by the 'ulama'.

Among the subjected communities, Manicheanism (a sect of the Gnostic
type founded in the third century under the first Sasanians) had never re-
ceived the protection accorded the more established religious allegiances.
Throughout the Roman and Sasanian empires it had been bitterly persecuted
under both Christian and Mazdean rulers, wherever it had spread. The same

20,0 THE CIVILIZATION OF THE HIGH CALIPHATE

attitude of execration was taken up by the Muslim 'ulama' as soon as they
became aware of the Manicheans' existence. They refused to admit that Mani,
the founder, who had lived two centuries after Christ, was one of the true
prophets; his followers, therefore, were not to be tolerated as were those of
Moses, Jesus, and even Zoroaster. Though for a time Manicheanism was able
to convert even kings in central Eurasia, in the end it succumbed to the unani-
mous enmity of the world's rulers. It is a unique example of a major 'world
religion' which has left no living traces.

Manicheanism was not a people's religion; it was not enshrined in the peas-
ant's heart. It represented a particular style of piety which long appealed to
spiritual seekers, including many of an intellectual bent (one of these was
Augustine). In the eighth century this sort of piety seems to have appealed
strongly to a great many Muslims; the tradition which carried it was not
limited to an endogamous group, as were the other non-Muslim traditions, but
played an active role in the course of events. It seems to have won many secret
followers at court among persons who were officially Muslims. The danger it
posed was at such a level that the earliest Muslim theology seems to have been
developed partly for the sake of opposing the Manicheans intellectually.

Manicheans saw the universe as being sharply divided between spirit and
matter. Spirit was free, creative, beautiful, true, and in general good; matter
was cold, destructive, evil, in general just the reverse of spirit. The human
soul was captive in a dark material body and could escape to the true Light
of Spirit only by rejecting that material body so far as in it lay, allying itself
with the positive forces of Spirit as they broke through into one's life. The
extremest asceticism was to be prized above all; if one could not be highly
ascetic oneself, one could at least associate oneself with those who were, for
in them Spirit was overcoming Matter. The Manicheans reckoned some of
their number as having achieved the spiritual goal of renunciation. These
were honoured and served by the rest, while ordinary individuals tried to
carry out at least a minimal curbing of the flesh.

Such an orientation fostered at once gentleness and aloofness. Manicheans
were known as being unwilling to kill even animals and as being ever ready
with acts of charity. On the other hand, they viewed the rest of mankind as
lost in ignorant darkness. They did not, indeed, push their search for moral
purity to the point of rejecting all imaginative culture, as have some puritans.
They had their own system of all-inclusive natural science, on which they
prided themselves. Even their enemies admired their school of magnificent
painting. But the sustaining base of their faith was its transcendentally as-
cetic mood. They rejected the world'of material existence loved by the mass
of people (including aristocrats), in favour of a truth into which could enter
only a pure elite (an elite of the spirit, of course, where earthly distinctions did
not matter).

For some time, apparently, a good many well-placed persons, intellectually
inclined, found in Manicheanism a spiritual counterbalance to the polished

THE ABSOLUTISM IN FLOWER 291

life that emanated from the court, to the clever brilliance of the litterateurs
and men of manners, the adibs. Its faith was more serious than the usual
courtly ideals; yet it seemed to go deeper than the legalistic system being
developed by the Piety-minded 'ulama', with their matter-of-fact populistic
spirit. The populistic tendencies of Manicheanism, with its recognition that a
person of any social status could join the elite, were tempered by its strong
sense of an esoteric truth not accessible to the uninitiated. Tending to exalt
detachment from mundane affairs, it no doubt appealed to world-weary cour-
tiers. But its very appeal to sophisticated Muslims earned it the hatred of the
'ulama.'. In refusing to grant Manicheans dhimmi status along with other
religious communities, they condemned them, in effect, to a mass choice be-
tween conversion and extermination; and the 'ulama.' were yet more indignant
at what may be called 'Manichean Muslims', who formally professed Islam
but adopted the Manichean world view. Al-Mahdi manifested his support of
Islamic purity in a practical way at the expense of this sect by killing any
courtier who would not or could not free himself of its taint.

In the ninth or tenth centuries, Manicheanism lost ground in Islamdom
and everywhere. The Arabic name for it, Zandaqah, became a word for every
sort of socially abhorred heresy; anyone suspected of cloaking an esoteric
faith beneath his profession of Islam was called a Zindiq. The original appli-
cation of these terms came to be forgotten.

By the end of al-Mahdi's reign, the bulk of the 'ulama' seem to have been
reconciled to the 'Abbasi compromise, and the absolute monarchy had gained
the minimum institutional religious support that it needed. Even most sup-
porters of either Kharijism or Shi' ism were at least willing to acknowledge the
de facto success of the 'Abbasi regime for the time being. In the latter part of
al-Mahdi's reign, the Egyptians made a bid for independence under a descen-
dant of ' Abd-al-' Aziz, the Umayyad governor whose family had long had prime
prestige in Egypt; but the hard core of his supporters turned out to be Khar-
ijis, and when he failed fully to accept their doctrine, the movement fell apart. 3
When upon al-Mahdi's death the 'Alid party again revolted in the Hijaz, it
was easily suppressed.

Hdrun al-Rashid: the caliph and his court as patrons of culture

Al-Mahdi's son Harun al-Rashid (786-809), after a brief reign by al-Rashid's
brother al-Hadi, enjoyed the position of caliph at the peak of its splendour

1 Abu-'Umar Muhammad al- Kindt (d. 961), who reports this movement in his Kitdb
umard' al-Misr, ed. Rhuvon Guest (Leiden, 1912), pp. 124-30, represents the care-
ful gathering of data — especially on a local basis (in this case, Cairo) — which is the
foundation of most of our knowledge of Islamicate history. Typically, al-Kindi is not free
from occasional legend, and he fails to see the wider issues which his details illuminate for
us, but his painstaking listing of names and remembered acts and qualities seems reason-
ably objective and in any case allows for extensive correlation and verification with other
sources of data. But his work would not be readily accessible to us without the pains-
taking modern editing of men like Guest.

292 THE CIVILIZATION OF THE HIGH CALIPHATE

and luxury. For a brief generation there was relatively unbroken peace and
prosperity in the caliphal empire. The occasional rebellions and foreign wars
were important only locally. Al-Rashid was recalled fondly as the ideal great
monarch in the stories of the Thousand and One Nights. His reign typifies,
indeed, the society of classical Baghdad at its height.

Under the caliph, the government was now largely delegated to adminis-
trators, a vizier as financial chief and generally head of government, and his
secretaries in their many bureaus (diwdns). The caliph was not expected
necessarily to take a personal role in government, but rather to be a court of
ultimate appeal. An energetic caliph, or a lazy caliph in an energetic mood,
might indeed make his own decisions: but commonly enough even matters of
general policy were left to the vizier. When this system was fully developed, a
caliph would change the line of policy not directly but by choosing a new vizier.
Al-Rashid tended to intervene chiefly in matters of his personal concern or
special interest, such as charities.

It was hoped, however, that a caliph would nevertheless perform two cere-
monial obligations of great weight. He should lead the Friday salat worship in
the capital on special occasions at least. In this connection the caliphs made a
point of acting as the heirs of Muhammad; they acquired a cloak the Prophet
had allegedly worn and other personal items of Muhammad's, in which they
could deck themselves at appropriate moments. (Al-Rashid and most who
followed him usually preferred, however, to leave the actual leadership to a
representative, themselves simply forming a part of the body of worshippers,
though safely apart in a specially partitioned area in the mosque, the maq-
siirah). The caliph should also lead the armies on the great jihad raids into
Byzantine territory, even though he left actual military decisions to generals.
(Al-Rashid preferred to alternate between leading the caravan of the hajj
to Mecca one year and the jihad army the next.) For it was not forgotten that
the caliph remained in principle the amir al-mu'minin, commander of the
faithful; managing finances and even dispensing justice he might leave to
others, but not the common Muslim undertakings, the public worship and the
war against the Tyrant — as the Byzantine emperor was called.

This conception of the caliph's role reinforced a concept of government al-
ready common enough in agrarianate times. That the ruler should be above
all the commanders in war had answered, as late as Marwani times, to a certain
individualism on the part of Arab tribesmen, jealous of their freedom. Over
against the wider population of the Nile-to-Oxus region it now implied an
attitude of social laissez-faire. Al-Rashid's son, who took the duties of govern-
ning very seriously, summed up those duties under three heads: maintenance
of justice in the courts, of security in streets and highways, and of defence on
the frontiers. That is, the government had no concern with positive social or-
dering, with actively creating conditions for the good life (this was left to
family tradition or to the efforts of the 'ulama', according to taste); it was
simply to guarantee security: security against fraud, against force by in-

294 THE CIVILIZATION OF THE HIGH CALIPHATE

dividuals, and against force by alien groups. If the central resources were used
for digging wells or even for building cities, this was looked on as private
benevolences by the caliph or the vizier; benevolences incumbent on them, no
doubt, but as rich men rather than as officials. The Arab and the Islamic
heritage encouraged people to conceive absolute government in its most in-
dividualistic sense, the role of common endeavour being reduced to a mini-
mum, the role of individuals (generally, privileged individuals) maximized.

However restricted his daily political role, in the affairs of high society the
caliph was actively leader of all. The caliphal family itself took first place, as
in receiving, so in disbursing wealth as patrons of the arts of luxury and of
learning. Ibrahim al-Mawsili (d. 804), the most celebrated musician, and
Abfi-Nuwas (d. c. 803), the most celebrated poet of the time, both of whose
lives were identified with wine and the gaiety of song, lived from the wealth of
the court, as was to be expected. A cleverly turned poem could win a bag full
of gold, a horse from the caliph's stables, a beautiful singing slave girl — or all
three at once. On every occasion of courtly joy, largesse was scattered among
the populace, rich and poor — in the form of coins tossed in abandon in the
streets, of food served to all comers, or of robes of honour of luxurious silks or
brocades passed out among favoured courtiers. A less spectacular but more
dependable way of rewarding talent or expressing favour was to grant an
individual the revenue for life from a given village, or to give him lands out-
right from the government's holdings.

Along with poetry and song, the courtly life was felt to be incomplete with-
out a large component of more sober learning and even piety. Learned literary
culture was represented in al-Kisa'i (d. 805), for instance, who was one of the
most eminent grammarians of the day and at the same time the tutor of the
royal princes. Some eminent 'mama.', indeed, refused to be found too close to
the caliphal court; but al-Rashid — between his bouts of wine-drinking — paid
attention to their exposition of the religious law. There are several anecdotes
of how Abu-Yusuf, chief qadi and the most prominent legist of the Iraq after
Abu-Hanifah, solved legal puzzles for the caliph in such a way that the letter
of the Shari'ah law was maintained; and whatever their genuineness, it is
certain al-Rashid and his court honoured the authority of pious Shar'i scholars,
favouring those specialized in hadith reports, and especially those of the
Hijaz. Al-Rashid tolerated their independent spirit within the limits of out-
ward public propriety.

Al-Rashid's whole clan, the 'Abbasids in general, but especially his immed-
iate family concentrated the largest spending in their own hands. His wife
Zubaydah made herself famous for her charities, notably causing numerous
wells to be dug along the pilgrim trail from the Iraq to Medina for the use of
the hajj pilgrims. Other families close to him might be almost as munificent.
The 'dynasty of viziers', the Persian Barmakid family, showered out almost
as many gifts as the caliph himself. As a group, the various big spenders of
Baghdad attracted artists, poets, scholars, philosophers, as well as tricksters

THE ABSOLUTISM IN FLOWER Zg5

and sheer career men from every part of the empire; whoever could please
one or another great man had his fortune made — provided he could avoid
a sudden disgrace and execution, such as befell the Barmakid favourite
one day.

The Barmakid family had been Buddhist priests at Balkh (near the Oxus)
before their conversion to Islam. Already under al-Saffah and al-Mansur,
Khalid al-Barmaki had become a top-ranking katib, secretary, and trusted
adviser of the caliph. Al-Mahdi had placed Khalid's son Yahya in charge of
al-Rashid's affairs when, as a youth, he was a titular provincial governor; al-
Rashid had grown up close friends with Yahya's son Ja'far, whom he retained
as chief courtier and companion of his leisure hours, making his brother al-
Fadl b. Yahya head of affairs during most of his reign. By 803 al-Rashid had
grown jealous of the enormous power of the Barmakids, to whom suitors for
favours naturally turned even more readily than to the caliph himself. One
night, without notice, he had Ja'far beheaded. (The head was publicly dis-
played next day as usual, to prove to his partisans that he was dead and
beyond help from any coup.) Ja'far's brother, al-Fadl, and their father were
imprisoned and died not long after.

Luxury was most brilliant at the capital — and probably most insecure
there. But the well-to-do could lead a very comfortable life even in the pro-
vinces. Nishapur, for instance, the capital city of Khurasan (founded under
the Sasanians — though presumably some such city had been in the vicinity
still earlier) was noted for its wealth and healthfulness. It was watered by
streams from the hills to the northward in the wet season (winter) and there
were famous garden districts around it bearing all kinds of fruits; the Nisha-
puris took a special interest in techniques of fruit tree breeding and cultiva-
tion. In the Arid Zone, the use of water was highly developed. Long under-
ground water channels (qanats), sometimes hewn through rock, carried water
from mountain sources far out over plateau or desert. Such means were used
at Nishapur not only for irrigating gardens and fields but also to provide water
for the city: every well-to-do household (and with it not only the master
family but all the dependents) had its own underground water source and its
cistern for the dry season. The water then ran out below to irrigate the fields.
Travellers were awed by its magnificent Friday mosque (the one most heard
of was built by the Saffari power subsequent to al-Rashid's time), which both
rich and poor enjoyed, and by the diversity on display in its markets as well
as its manufactures.

Cities were not alike. In contrast to the healthfulness of Nishapur, we hear
of the filth of Gurganj in Khwarazm, in the delta of the Oxus (as contrasted
to the Khurasan plateau): travellers complained that though the city was
wealthy, yet there the offal was left in the streets so that even the mosque
(generally kept fairly clean by removal of shoes at the gate) was dirtied by the
feet of those who entered without having worn shoes outside. Gurganj may
not always have been so filthy, but it did not have the drainage system of

296 THE CIVILIZATION OF THE HIGH CALIPHATE

NisMpur. However, in any wealthy home there would be many amenities:
rugs and pillows to sit on, elegant book stands, carved chests and handsome
jars for storage and good looks alike; low cushioned daises and pools in court-
yards, often with at least a miniature waterfall; braziers for warmth in winter
and water evaporation for cooling in summer. Perhaps such luxuries as ice in
summer could be obtained more readily at Baghdad than in most provincial
towns, but the privileged classes of the empire lived a life comfortable
and delightful; and if they were subject to the vagaries of natural calami-
ties, disease, and occasional riot, they were relatively free of the threat of
war.

Letters and learning

The point of greatest splendour in the caliphal court was its patronage of let-
ters. These were, above all, Arabic letters: ever since the administration had
been put into Arabic, the traditions of a cultivated bureaucracy were
being built up in terms of that language. Formal Arabic prose received its
first great impulse with the translation into elegant Mudari Arabic of several
Pahlavi works. (Later, at least, certain forms — such as lives of saints — seem
to have been based on Syriac traditions also.) Among translations from Pah-
lavi, most notable was a collection of tales {Kalilah wa-Dimnah, anecdotes of
two jackals as advisers to the lion-king) ultimately taken from the Sanskrit
(allegedly in Nushirvan's time). The translator was Ibn-al-Muqaffa', the ex-
Mazdean administrator who had urged al-Mansur to a more systematic ab-
solutism. This prose was limpid, entertaining, and edifying; for some time
such remained the norm of the Arabic prose honoured by the katib class, the
courtiers and bureaucrats who supported the absolutism. They delighted in
literature which was worldly-wise and informative and could add to the polish
and brilliance of their sophisticated conversation — and of their official corres-
pondence. The adab, the polite cultivation of that class, gave a large place to
verbal brilliance and hence to literature.

But at least as important as prose, for the adib and for the katibs generally,
was poetry. But poetry is essentially untranslatable; in this sphere, though
the tradition of Sasanian administrative culture may have established the
literary status of poetry, its substance at any rate came from the old-Arabian
tradition, to which the katib class was early and enthusiastically committed.
Here too, however, a certain effort at mental translation was required. For
whether in its pre-Islamic Bedouin form or in the form it took when revived
in the Arab garrison towns of Marwani times, that tradition was alien not
just to the family heritage of most of the Muslims of 'Abbasi times, nor even
to the daily language, which was (if Arabic at all) an eroded, 'settled' Arabic,
and the old inflected Mudari of the Bedouin. More important, the Arabic
poetic tradition was alien also to the deeply urban patterns of the bureaucracy
and of the other elements in the population which now turned to honouring

THE ABSOLUTISM IN FLOWER 297

it. Hence among literary scholars (like al-Kisa'i) arose a school of collectors
and editors of the old poetry, dedicated as much to its philological niceties as
to its aesthetic delights; while the more poetically gifted, rebelling against
the limits which interpreters of the tradition had thought to lay down for it,
took up the sense of Arabic rhythm and image and the straightforward spirit
with which the tradition confronted them and transmuted all that into a
form and a mood more appropriate to the courtier, the katib, and even the
merchant. This process had indeed begun under the Marwanids, but was now
fulfilled.

Abu-Nuwas of Basrah (and of Persian stock) was personally a libertine and
dedicated his verse to love and wine. He had studied philology at Kufah and
was said even (as was the custom for students of the language) to have spent
some time in the desert with the Bedouin; he glorified licentiousness with an
echo of such old-Arabian models as the love-prologue that initiated a Bedouin
formal ode (qasidah); but he rejected the heroic grandeur that had gone with
the qasidah in favour of a more intimate, even a pert and playful, relaxation.
Abu-Nuwas dedicated much of his erotic verse to the love of youths, thus set-
ting a fashion that was later to become fixed in some Islamicate circles even if
the poet had no personal homosexual interests. AM-l-'Atahiyah (d. 826),
though of Arab stock, went even further in transforming the poetic tradition:
ascetic and of a speculative turn of mind — yet a loyal Muslim of Shi'i senti-
ments — -he dedicated his verse to a philosophic melancholy which proved more
popular in the market place than at court; his lines formed a mine of soberly
pious quotations for future popular writers.

Along with the cultivation of literature went a reflection on the mechanics
of that literature. A major achievement of the time was the systematization
of Mudari Arabic grammar (which subsequently became the model for gram-
matical analysis in other Semitic languages). Such analysis had already begun
in Marwani times. After the experience of two generations, Sibawayhi of
Basrah (d. 793), representing a tradition more faithful to the vagaries of
actual usage than were some Kufan grammarians, produced the standard
book of grammar (afterwards called simply The Book) — though naturally he
left many minor problems to be worked out by later generations. The task,
of course, was to sort out the ways in which various particles, word orders,
verbal or nominal patterns, etc., could be used without causing confusion
through conflict with other usages. It was not conceived in just this functional
manner, indeed; grammarians saw themselves as identifying a complete set
of the natural distinctions among types of word and of sentence pattern, and
then discovering, from ancient poetry, from surviving Bedouin usage, and
from Qur'an and hadith, how each type had in fact been used among respect-
able Arabs of Muhammad's time. But living speech within a given generation
does, in fact, ordinarily present mutually consistent usages. At any rate, the
result was to establish standards which effectively safeguarded essential intel-
ligibility, however specialized the jargon became, however complex the sen-

20.8 THE CIVILIZATION OF THE HIGH CALIPHATE

tence — or however alien to the old Arabic by birth was the person under-
taking to compose in it. No other Islamicate language was ever so minutely
examined and disciplined.

Already under al-Mansur, the caliphal court had come to patronize, beyond
Arabic belles-lettres and the work of Shari'ah-minded piety, learning of the
most varied sorts. In contrast to Kilfah and Basrah, which long remained
important centres of the Arabic and Shar'l studies, Baghdad thus became a
centre not only of these studies but of natural science and metaphysics. The
most famous works of medicine, mathematics, and astronomy, especially,
were translated: probably first from Pahlavi and Syriac but then also from
Sanskrit and Greek. Whereas the Arabic and the Shar'i studies were pursued
chiefly by men born or converted to Islam, those who pursued natural science
tended to retain their older religious allegiances as dhimmis, even when doing
their work into Arabic. If their work found favour at court, it was not so
much as direct expression of the cultural ideals of the katibs but rather as
having practical personal use or else as fulfilling individual speculative curi-
osity. Al-Rashid's son, al-Ma'mun, was to prove personally much interested
in the science and even the philosophy of the Hellenistic tradition; by his
subsidies and encouragements he gave a great fillip to the developing move-
ment of translating from Greek and Syriac into Arabic the classic works of
medicine, astronomy, mathematics, and natural philosophy generally. Al-
Ma'mftn endowed (in 830) a fine research library, the Bayt al-Hikmah, 'House
of Wisdom'. He went so far as to seek for manuscripts in Constantinople,
where the Greek tradition was naturally highly cultivated. From his time on,
the quality of the translations improved greatly.

But the most important centres where students could gather such learning
were older cities within the empire. It was from the great medical school
founded in Sasanian times at Jundaysabur in Khuzistan that the Islamicate
medical tradition was to get its greatest impulse. Already al-Mansur had
brought from there to Baghdad a prominent Nestorian physician (Bakht-
Ishu'), who founded a line of physicians in the capital. The town Harran, in
the Jazirah, had become perhaps a still more important centre. It had to some
degree succeeded Alexandria and even Antioch as an active centre of learning.
In Harran, even Christianity had not greatly penetrated and older Pagan
cults had been perpetuated; in Islamic times the Harranis called themselves
Sdbi'ans, a term found in the Qur'an to designate an otherwise obscure group
of tolerated monotheists, and so managed to pass as possessing a legitimate
form of prophetic religion; the neo-Platonic (or neo-Pythagorean) faith of
their educated class, indeed, justified this. Sabi'ans, Christians, and Mazdeans
increasingly taught Muslim students and used the Arabic language. With the
patronage of al-Ma'mun, Baghdad soon became the greatest centre of such
science and philosophy in the empire, as it had become that of Arabic belles-
lettres and of Shar'i Islam.

THE ABSOLUTISM IN FLOWER 299

The war of succession to al-Rashid

Wealth and power from the eastern Maghrib to Sind and the Oxus basin were
concentrated almost uprecedentedly in the splendid families at the heart of
the 'Abbasi state in Baghdad, which Shar'i 'ulama', Arab tribesmen, Persian
landlords, cosmopolitan merchants, and (more passively) peasants of every'
land were united in regarding as at once the symbol of Islamic unity and of
civil order and powerful justice. The acquiescence of the Muslim idealists and
the relative satisfaction of traditional Irano-Semitic political expectations
seemed to assure the 'Abbasis a solid power. The former Iraqi centres, Kufah
and Basrah, where law and grammar had developed, increasingly yielded their
sons and their students to the attractions of the capital, where all viewpoints
mingled.

Al-Rashid himself made a move which turned out to illustrate how unsure
this caliphal state power was as compared, say, with that of the Sasanians of
yore. He decreed (and duly caused the Muslim notables to agree) that the
control of the empire was to be divided on his death: the Fertile Crescent and
the western provinces were to go to his son al-Amin (with the dignity of
caliph) while Khurasan and the eastern provinces were to go to another son,
al-Ma'mun, with an army and full autonomy (and the right of subsequent
succession), though al-Ma'mun also was bound to obey al-Amin in ultimate
matters. Al-Rashid posted his decree in the Ka'bah to give it sacred authority.
There was no one social element in the state strong enough to gainsay this
dismemberment of the hard-built political structure for a personal whim.

This social fact was reflected in the legal situation. There was no basis for
theoretical protest by the Jama'i-Sunnt 'ulama', who did acknowledge the
legality of the 'Abbasid caliphate. In accepting the 'Abbasid dynasty for the
sake of Muslim unity, they had grudgingly accepted, for want of a more
principled norm, the dynastic rule already in use among the Marwanids: de-
signation by the predecessor, confirmed by public acceptance among the
notables of the chief cities. But the 'ulama', refusing to recognize the absolutist
state and wanting to deal even with public duties as personal responsibilities,
before God, of those who happened to be charged with them, interpreted that
designation and its acceptance as purely personal convenants, not as autono-
mous state institutions; hence the designating caliph could attach to his
designation such conditions as he chose — including a subsequent succession —
and once these conditions had been accepted and sworn to by the designees
and the notables of the Muslim community, the only Shar'i legal principle
that applied was that of fidelity to covenants. But if the 'ulama' had no basis
for protest, neither had the katibs themselves, for all their direct interest in
the monarchy. To supplement the Shari'ah, there was no generally recognized
dynastic or monarchic body of law, based on the nature of the office as such,
to override such conditions: no rule of primogeniture, for instance, nor any
definition of the rights of an office as such, which would have a prior claim on

300 THE CIVILIZATION OF THE HIGH CALIPHATE

men's consciences and allow them to brand al-Rashid's covenant as illegal
even apart from Shar'i principles.

For some years after al-Rashid's death (809 — in the course of an expedition
against Kharijis in Khurasan) the empire was in fact divided. The division did
not then last, however — a civil war soon ensued. The warring lasted for over
two years (811-813) and a number of factions participated, each aligned
against all the others. Yet the struggle took on a very different air from the
Second or the Third Fitnah wars, when diverse viewpoints, each appealing to
all Muslims, brought forward their candidates for the caliphate. The main
struggle was between al-Rashid's two sons, each of whom (or his advisers)
desired to be supreme in the whole empire and believed this possible — though
the immediate occasion was the question whether al-Amin could set aside al-
Ma'mun, who was to succeed him if al-Amin died first, in favour of al-Amin's
own son. The choice was partly one of personality: al-Amin (favoured by his
well-born mother, al-Rashid's wife Zubaydah) had the reputation of being
debauched and hence manipulable by his ministers, while al-Ma'mun (son of a
Persian slave-girl who had not survived to have even such influence as slave-
mothers could have) was intelligent and more likely to be independent. How-
ever, al-Amin's incompetence and the victories of al-Ma'mun's generals soon
won some even of al-Amin's ministers, who could have expected to control
their ward, over to al-Ma'mun. Yet Baghdad held out against al-Ma'mun's
besieging general Tahir f or over a year despite fire and famine and lack of
support from the provinces. Al-Amin had the support of the great families
and of the populace in the capital, who evidently feared the Khurasani
orientation of al-Ma'mun; for al-Ma'mun kept his headquarters at Marv in
Khurasan and had Persians as generals (such as Tahir) and as ministers. Al-
Rashid already had preferred Raqqah on the Euphrates to Baghdad as his
headquarters and possibly the Baghdadis feared for the status of their city, in
which al-Amin had spent the imperial revenues profusely. In any case, no
Islamic principle seems to have been at stake.

Even outside Baghdad, the accompanying struggles tended to be of local
scope. During al-Rashid's reign an 'Alid rebel had held his own for a time in
the sub-Caspian mountains in the Shi'i cause, and Kharijis had led a major
rising in the Jazirah, which was put down; and Kharijis were again active in
Khurasan during the war of succession. Shortly after al-Ma'mun's victory,
the Shi'is of the Iraq made a potent bid for power. But during the war of
succession itself, neither Shi'is nor Kharijis played a major role. In Syria,
Damascus made a bid to regain the caliphate, but factional fighting between
the Kalb and the Qays tribal blocs there ruined the chances of their candidate.
In Egypt, opposing local factions took advantage of the central government's
weakness to attempt each to master the other, but the theoretical supremacy
of the 'Abbasid caliphs was not questioned.

At last when Baghdad was crushed, a single ruler was recognized again
throughout most of the empire of al-Mansfir and al-Rashid. Al-Ma'mun was to

THE ABSOLUTISM IN FLOWER 301

face many rebellions, but he was the undoubted master of most of the empire
most of the time. For a season the wondrous prosperity of the great monarchy
was prolonged. On the basis of this prosperity, the new Islamicate culture was
being created.

Economic expansion and the popularization of Islam

Islam became a mass people's religion on a wave of economic expansiveness;
which, indeed, carried the whole range of cultural innovation of which Islam
was a part. As the commercial activity along the interregional trade routes of
the hemisphere increased, the tendency to intensive urbanization in the region
from Nile to Oxus increased. Baghdad, a major trade centre even when the
caliphs fled from it, became larger than had been Seleucia and Ctesiphon, its
Hellenistic and Sasanian predecessors. This urban economic expansiveness, so
far as it was a product of hemisphere-wide economic activity, contributed to
making possible the power of the caliphal state; but it was in turn accelerated,
at least, by the absolutism itself.

As generally happens, economic expansiveness tended to be self-perpetua-
ting. The surging urban prosperity allowed fortunes to be made by large num-
bers and encouraged a social mobility rapid even for the Nile-to-Oxus region.
Under the broad peace established by the empire, commerce flourished freely
and markets, and hence possibilities for new investment, were expanding.
Merchants, buying cheap and selling dear, found it profitable to carry exotic
goods long distances with less danger of interruption and consequent increase
of price than in politically more troubled times; hence luxuries were available
not only at Baghdad but in many other places in unusual abundance. Such
trade stimulated further local production. Of the wealth so released, at least a
part went to encourage still further economic activity. To supply the needs of
far-flung merchant houses, a regular banking business grew up, which evi-
dently had developed, as often in agrarianate times, out of money-changing
business. Bankers accepted draughts drawn in one place on funds deposited
far distant. The government naturally made use of these bankers in expediting
its tax collecting and in other operations of its bureaucracy; as we have noted,
without the commercial network the effective bureaucratic centralization
could hardly have functioned. But conversely, without the dependability pro-
vided by the bureaucracy the commercial network would not have been able
to function in such large-scale units as it often did.

Though money still often went into land, the most dependable source of
income in the Agrarian Age, funds available for investment were very com-
monly put into trade. Transport of luxury goods was most spectacularly
profitable, but less consistently so than transport of more mundane goods like
grain, which was also important. Commercial arrangements always tend to be
relatively flexible, but this was a time of unusual diversity and inventiveness.
Almost all forms of mercantile life tended to be carried on in partnerships;

302 THE CIVILIZATION OF THE HIGH CALIPHATE

these varied almost indefinitely as to the terms of co-operation between the
partners and as to the number of partners engaged in a single endeavour.
Muslims and non-Muslims were sometimes partners together; this could be
convenient: for instance, in a partnership between a Muslim and a Jew, the
Jew would take off on the Sabbath and the Muslim on Friday. To maintain
the far-flung contacts which long-distance trade required — a merchant sta-
tioned in Gujarat (in India) might have a relative travelling in Spain or be
planning a commercial venture to the Oxus basin or the Volga valley — there
were regular postal services, arranged by private individuals, over the most
important sea and land routes; usually these were slow, but some seem to
have been express services; perhaps, as in the official messenger services
{band) for central government use, messengers could carry missives very
rapidly by changing horses at pre-arranged servicing stations. 4 The merchants
themselves travelled with their goods in ships — normally owned by private
individuals, whether on seas or on rivers — or in privately managed land con-
voys of camel transport, the caravans.

Trade was the usual investment, but some funds were invested in hand
industries. A number of new industries spread in the region under the High
Caliphate: paper-making, learned from China via the Oxus basin, replaced the
processing of papyrus leaves for writing material. The use of cotton came in
from India. Paper and sugar mills sometimes employed large numbers of
workers — as did government enterprises such as the workshops in which were
made the robes of honour which the caliphs dispensed. Nevertheless, the ideal
of craftsmen was to work as independent masters at their trade, perhaps hiring
one or two other workers as assistants; or to work in apprenticeship, if
they could not afford to set up shop on their own. Both merchants and
craftsmen found ways of associating to protect their common interests against
outside competition; among dhimmis, this could be done through the com-
munity organization of the local dhimmi group itself. But at least in some
places — like Cairo — individuals tended very often to be almost on their own.
For instance, though commonly tradesmen of the same calling were located in a
common market area, sometimes individuals took space elsewhere ; and though

1 Cf. S. D. Goitein, 'The Commercial Mail Servicein Medieval Islam', Journal American
Oriental Society, 84 (1964), 1 [8-23, which reports on those mail services that led to
Cairo. The Arabic word barid meant 'official messenger service'. For 'postal service',
maintained privately for private use, there were quite different terms (e.g., fayj).
Unfortunately, a mistranslation has occurred which has occasioned a fair amount of
wasted ink. The French term 'courrier' has been quite legitimately used for the word
barid, in one of the several meanings of 'courrier'. But since 'courrier' can also mean
'postal service' as well as 'official messenger', readers have tended to be misled and have
had to be warned that the barid did not carry private letters. In English the confusion
is quite inexcusable; the use of 'post' for 'courier' is quite archaic. But unfortunately
early scholars wrote as if translating not the Arabic barid but the French 'courrier',
and mistranslated it; so that the barid is commonly miscalled 'postal service', with
which it had little in common; then every time the reader must be warned at length
that it was not, in fact, a postal service. The French confusion has entered the modern
Arabic, in which barid does in fact, mean 'postal service'.

THE ABSOLUTISM IN FLOWER 303

the tradesmen of a given trade were often referred to collectively on a family
basis, individuals could frequently take up a new trade which their families
had not pursued. The principle of free contract ruled in a wide range of
city life, allowing for much fluidity such as an expanding economy
required. 5

In this expanding prosperity and its concomitant social mobility, Islam
and the Muslims' institutions played a key role. To a degree, the monarchical
discipline and broad extent of the caliphal state — under strong caliphs, at
least — had its effect in assuring tranquillity and a favourable atmosphere for
trade. With order came prosperity. To a degree, of course, it was just the
prosperity which made possible the strong rulers. But once the monarchy
arose, it had a further autonomous effect. In particular, the extensive power
of the monarchy had its own direct influence in producing a ferment in high
culture. Even after the caliphal monarchy had been destroyed, the new
questions raised and the new traditions launched bore fruit in a vigorous
continued cultural dialogue.

The court was the focal point of the prosperity of the times, and the ex-
ample and influence of the court spread throughout the empire because of the
attractive power of wealth. Whatever of the revenue — that is, of the taxes —
could be gotten away from the provinces came to Baghdad, there to be re-
distributed through the channels of a fashionable life of luxury; thither came
likewise, accordingly, the most ambitious young men from everywhere. From
the capital poured back into the provinces not only governors, but merchants,
landowners endowed from the caliphal bounty, and all sorts of men who had
occasion to taste temporarily of the splendour of Baghdad. The fashions set at
the court in administration, in social life, and in literary and artistic taste were
thereupon spread to all the provincial centres, while the court itself was a
melting pot in which all cultural traditions confronted each other.

A key trait of the new common culture so formed, one which helped in turn
to make that culture possible, was a high degree of social mobility: not merely
the social mobility in economic life presupposed by economic expansion, but a
specifically cultural openness — based on the possibility that a man of spirit
or of special gifts could rise in the social scale without the advantages of
family or of communal connections, or could move among circles formed by
other communities despite the advantages which local ties gave him in his
own. The Muslims were aware of the vitality which could come from 'new
men'. We read of a debate between two scholars, one of whom came from

5 The articles of Professor Goitein about the manuscripts in the Cairo Geniza, in his
Studies in Islamic History and Institutions (Leiden, 1966), have thrown much light on
conditions of trade, especially among a minority group in Cairo. His work also includes
a good bibliography on Islamicate commerce generally. We cannot be sure that equal
liberty of trade prevailed in smaller towns, but it is clear that even there no hard and
fast guild structure inhibited initiative; cf. Ira Lapidus, Muslim Cities in the Later
Middle Ages (Harvard University Press, 1967), on Syrian towns in the (later) Mamliik
period.

304 THE CIVILIZATION OF THE HIGH CALIPHATE

Kufah and despised the other for coming from a little-known small town;
said the other, You degrade your city, while I bring lustre to mine. A man's
illustrious genealogy should begin with himself rather than end with himself.
Men of arts and letters were constantly on the move, rarely residing in the
town of their birth, and usually able to report visits to many widely distant
cities.

The caliphal court itself soon took the lead in encouraging social mobility.
The courtiers clung to some elements of hereditary position, which had been
very important in Sasanian times. Sometimes they wrote as if this were
essential to adab, to polite cultivation. The landed gentry continued to look
to noble descent — especially from royal heroes of ancient Iranian times — as
the criterion of social eminence. The descendants of the old Arabian tribes
still prided themselves on the purity of their line; and particularly the 'Alids
were generally looked up to, and the 'Abbasids themselves. (There were special
governmental bureaus to register the generations of these families, so that
genuine descent could be guaranteed, and to regulate the considerable pro-
perties and privileges accruing to the two families as such.) But such aristo-
cratic tendencies were at a disadvantage. Already in the time of Nushirvan,
the Sasanian nobility had lost some of its power. The Piety-minded among the
Muslims had further helped to lower the pride both of the Arab tribesmen,
with their pagan ancestors, and of the formerly infidel gentry. The 'Abbasi
programme was largely based on the principle of equality among the Muslim
faithful, even apart from the other sort of equality that prevailed among sub-
jects of an absolute king. At Baghdad under the 'Abbasi absolutism a man of
talent could make a fortune or reach the highest positions with little regard
for his birth. What mattered even at court was his personal adab, especially
his literary cultivation.

But it was not merely the economic and political role of the Muslim court
that made for a new common regional pattern of culture. Other social pres-
sures favoured by economic expansiveness expressed themselves directly
through Islamic religion. Its spread altered the social relations holding among
the several traditions of high culture from what had obtained in Marwani
times. By now Muslims no longer formed merely a small ruling class, but were
becoming a substantial part of the population at large, in some places a
majority. Consequently, there were no longer simply many religious communi-
ties divided from each other in geography, in language, and in lettered tradi-
tions — a situation which even the Mazdeans, linked to a specifically Iranian
ethnic sense, had not transcended. There was now available a community
which fit into and even fulfilled many of the expectations of the Irano-
Semitic tradition, but which also combined a neutral appeal, not linked to any
of the major agrarian ethnic groups of the region, with political power which
united all the lands of Iranian and Semitic heritage on a territorial scale which
the Sasanians had never matched. Even those who did not adopt the religion
tended to use the language of the new community; but the community won

THE ABSOLUTISM IN FLOWER 305

its position above all through the large number who entered into its religious
tradition.

Since the time of 'Umar II, conversion to Islam had been officially fostered.
Under Hisham, for instance, an application of 'Umar II's principles in Khura-
san and the Oxus basin had encouraged widespread conversion, undermining
the position of the old-line local rulers and contributing to the political fer-
ment which led to the 'Abbasi triumph. With the advent of 'Abbasi rule,
conversion seems to have become very general. After several generations, the
majority of the town populations and even many peasants in the greater part
of the empire seem to have become Muslims.

To be Muslim carried with it great prestige and no doubt merchants found
it a special advantage. Certain trades seem to have become largely Muslim,
while others remained largely Christian or Jewish. Above all, immigration of
country population into the growing towns (whereby the town populations,
never self-reproducing, were recruited) was probably often accompanied by
conversion to the dominant religious tradition there; the peasant who must
learn a new, urban, way of life anyway may as well learn a new religious pat-
tern too. Once the number of Muslims became considerable, the mosque be-
came the most lively, and certainly the most cosmopolitan, centre of all
activities. Popular story-tellers (qussds) held forth there upon the wonders of
the prophets and other Biblical and Talmudic tales, all conceived within an
Islamic framework. A popular Islam thus arose, complete with its own history,
ethics, and eschatology, drawing on the most striking notions found in all the
earlier religious traditions. When the towns had thus become primarily Mus-
lim, their innumerable links with the dependent village populations assured
the gradual conversion of the villages also.

Under the 'Abbasis, the new converts no longer had to be identified with
any Arab tribe, for such tribes themselves gradually were losing their privi-
leges. Islam became a badge, not of a ruling class, but of a cosmopolitan,
urban-oriented mass; it became a symbol of the newly intensified social mo-
bility. Of course, it was not only within Islam that social mobility showed its
effects. With accentuated urbanization came accentuated social mobility
generally. Many Jews who had been agriculturists in the Iraq turned to
commerce and became great merchants. But it was especially as Muslims
that persons took up more profitable careers or moved from one area to
another.

The piety of the dhimmi peoples

As the more active population, urban and urban-oriented, turned to Islam,
the social and cultural role of the older traditions shifted. Their people were
becoming minorities in the empire as a whole. But for a long time the rites
and folkways of these minorities — -which locally might still be majorities —
played a part in the life of the whole community which was not necessarily

306 THE CIVILIZATION OF THE HIGH CALIPHATE

creative but which maintained a continuity with the soil and with the past on
the level of daily and seasonal life patterns, which Muslims as such, with their
urban orientation, tended to lack.

At the time of the Arab conquest, the several religious groups which were
to form protected, dhimmi, communities gained equality of status among
themselves and protection against each other's interference; for instance,
against proselytism from one dhimmi community to another, which the Mus-
lims discouraged. Jews found a much more favourable position than they had
had under the Byzantines, at least; it was not only economic but political
openness which permitted an increase in their commercial activity. Even the
position of the relatively favoured Nestorian Christians temporarily improved
over against the Zoroastrian aristocracy, at least until the aristocrats turned
Muslim. The Monophysite Christian establishments of Egypt, Syria, and
Armenia profited immediately and immensely from the withdrawal of the
privileged Greeks. Quite apart from a general championing of piety as under-
stood in the Abrahamic tradition, the Muslims seem to have favoured the
viewpoint of the Monophysites on particular issues — notably the relative re-
pugnance to be found among Monophysites to religious statues and to figural
art, which perhaps sprang partly from their more general hostility to the
richer churches which could better afford such attractions. Already in
Marwani times, the Muslims began to frown on figural art in any religious
connection, to the delight of some Christians.

But by 'Abbasi times, the dhimmi communities were becoming isolated, as
communities, and as distinctive spiritual traditions, from the spiritual life of
the majority, at least on the level of high culture. They were becoming identi-
fied with individual ethnic groups. When we speak of ethnic groups, we mean
not nationalities as such, of course, but any groups with a common cultural
affiliation into which individuals are born, and in particular those smaller,
more cohesive groups that have a common language or dialect and a sense of
common loyalty as against outsiders, though they may not be living in a
single homogeneous area. Religious communities between Nile and Oxus had
long tended to be identified with such ethnic groups, and now the identifica-
tion became more rigorous. Almost every ethnic group that did not adopt
Islam came to be identified by its own special religious allegiance even more
than by its language. Thus Armenians, though they possessed a distinctive
language, were more certainly identified by their own church, which remained
independent not only in the Armenian mountains, where Armenians were
most numerous, but in Syria or Egypt (despite the similarity of the Armenian
theology to the Monophysite theology received there) or in the Iraq or Iran.
An Armenian who spoke Arabic was still an Armenian if he held to the church;
while an Armenian who abandoned the church (normally for Islam) soon
found himself a member of a completely different community. (Within Islam,
of course, old ethnic lines persisted on the basis of language or dialect, or of
locality or economic function, and new ones eventually arose, often associated

THE ABSOLUTISM IN FLOWER 307

with new Islamic sects; though generally such ethnic groups were not quite so
rigidly defined as were the dhimmi groups.)

The piety of each of the dhimmi religious bodies naturally retained its
distinctive character. Christians looked to the Redeemer who had suffered in
his love of human beings and had triumphed over death. Jews maintained the
sacred covenant, looking to the ultimate restoration of God's faithful people
and the purification of the world. Mazdeans strove to serve the triumph of
Light and Truth in a world where darkness seemed very strong. Such con-
tinued development in the forms of piety as took place in the several groups
was limited, however, to the internal life of a given group. It seems to have
had little effect on the relation of the various groups to Islam or to each other.

On the contrary, the piety developed in Islam may have had a noticeable
effect on that of some of the dhimmi bodies. There is some evidence that the
leaders of the Mazdeans'', in the eighth and ninth centuries, were partly re-
formulating some expressions of their tradition in response to the Muslim
impact. At any rate, the court Mazdeism of the Sasanians, which had allowed
a great deal of leeway in worshipping godlings alongside the Good god Ahura
Mazdah, seems to have quite disappeared. The priests now saw to it that
Ahura Mazdah became the centre of all devotion and a proper analogue to the
Allah of the Muslims. They bolstered this approach, of course, with theological
metaphysics and possibly even with reinterpretations of scripture. Whatever
reformulation took place, however, seems not to have hindered the rapid
numerical decline of Mazdeans relative to Muslims.

The older religious traditions made their contribution to the spirituality
which now took form within Islam, individual by individual and village by
village rather than as established communities. But throughout the High
Caliphal Period, this contribution was very important. We have noted that in
law and ethics and popular lore, which concerned everyone, both the 'ulama'
and the story-tellers of the mosque drew on the dhimmi background for their
spirit and often for their materials; the like happened in more special cases,
also: in the techniques of the ascetic and the mystical life, which only a few
concerned themselves with, and in the invention or perpetuation of local
cults. Indeed, as long as the dhimmis remained sufficiently numerous, they
remained the custodians of whatever rites were publicly acknowledged, in the
folk traditions of every day, for relating human beings to the round of the
seasons and to the local landscape. This happened all the more readily because
the Islamic lunar calendar — designed by soldiers and travelling merchants —
disregarded the solar seasons altogether; so that religious festivals on the
calendar appeared in every natural season in the course of thirty-three years,
and all festivals associated with nature had to be held by non-Islamic calen-
dars, inevitably associated with the older cults. Its cultic calendar branded
Islam as radically urban, detached from any given locality with its local
rounds of climate; the whole cycle of nature was abandoned to the older
traditions.

308 THE CIVILIZATION OF THE HIGH CALIPHATE

Even in the cities, a great many persons were not Muslims; but the great
non-Muslim strength was among the peasants. Except among the Arabian
(and Berber) tribes, to a large degree the people of the countryside at first
retained allegiance to their old traditions; and these country people were of
course far more numerous than the city people. Of the chief non-Muslim pro-
tected communities — the Christian, Jewish, and Mazdean — Christianity and
Mazdeism rapidly lost their position of social leadership as the landlords and
merchants turned to the ruling community. These two traditions particularly
became typical of the backward masses, especially but not exclusively of the
peasants, who were socially passive except in the case of the popular revolts.

From the Muslim point of view, at least, the most prominent feature of their
piety was its peasant emphasis on sacred shrines and on seasonal holy days
and festivals. In Syria and Egypt the whole population, Muslim and Christian
alike, celebrated the return of spring with the Easter holiday, when the
Christians had their processions while the Muslims looked on. In the same
way other saints' days marked the whole round of the agricultural year. At
the shrines dedicated to Christian saints, the peasants' hopes and fears ex-
pressed themselves in ways often directly traceable back to local pagan shrines
that had served the agriculturists of generations before. Eventually, many of
these shrines, and even the seasonal festivals occasionally, were to find a place
within Islam. For the time being they remained Christian; the Muslims had to
content themselves with a more or less vicarious participation in the age-old
piety of the peasant, with its superstitions on the one hand and its grand drama
of the returning year on the other.

In the Iranian plateau it was the old Iranian seasonal festivals, hallowed
by Mazdeism, that correspondingly persisted. They were celebrated on the
Sasanian solar calendar and, while they gradually lost their association with
the older religion, they could never be strictly Islamized.

West Mediterranean Islamdom

Already in High 'Abbasi times, the areas remoter from the centre of the
caliphal government were evolving a distinct historical pattern — extending
even to local relations between Muslims and dhimmis. This was especially true
in the largest distant region — the Muslim lands of the western Mediterranean
basin.

Most Muslim provinces of the west Mediterranean had never been subjected
to 'Abbasi rule at all. There the Berber population, converted en masse as
tribes and assimilated juridically to the Arabs from the start, had played the
part that Arabs had played elsewhere in Islamdom; but with acceptance
of Arab cultural leadership. The Maghrib, islanded between Mediterranean
and Sahara, was to the Berbers what Arabia, of a like extent (though mostly
less well-watered), was to the Arabs; but among the Berbers in their moun-
tains, pastoralism was more closely joined with agriculture. The Romans had

THE ABSOLUTISM IN FLOWER 309

played a larger part in the cities of its northern coast than had the Sasanians
in eastern and southern Arabia. Yet these cities, at least once the power of
the Roman empire was withdrawn, had little power over a hinterland which
had not been reduced, for the most part, to peasant status except in the
immediate vicinity of the cities; there was no network of secondary market
towns in the Maghrib to form a base for extensive citied agrarian domination.
Within the Maghrib, Islam seems to have become largely a matter of
slogans under which relations with outsiders could be regularized. The Berbers
had crossed into Spain in much the same spirit as that in which the Arabs
had come into the Fertile Crescent.

In Spain the Berbers who had come over were an unruly governing
class alongside a still more limited number of Arab families. Their feuding had
been held under a minimum of control by the caliphal governors. When the
Syrian caliphate foundered, a young member of the ruling Umayyad family,
'Abd-al-Rahman, escaped the massacre of his cousins and after numerous
adventures arrived in Spain, where he was able to persuade the diverse groups
among the ruling Muslims to accept him as arbiter under the title of amir,
commander, instead of a governor sent by the upstart 'Abbasid. He and his
successors managed to maintain a precarious supremacy for over a century
and a half, supported sometimes by a new bloc of Arab families from Syria
who set themselves off both from the Berbers and from earlier Arabs who had
come via the Egyptian Maghribi advances. (In the tenth century one of
'Abd-al-Rahman's scions transformed this Umayyad amirate into an
absolute rule as a caliphate modelled on that of the 'Abbasis.)

Spain's Latinized population, remote from the main centres of civilization,
had been ruled before the conquest by an aloof Germanic aristocracy and a
rigid church hierarchy which combined to repress any intellectual or civic
stirrings. Most of the cities were readily at the disposal of the new, more liberal,
rulers, who had allowed the desperately persecuted Jews their freedom and
left the Christian population to their local Roman institutions. But neither
Berbers nor Arabs were then able to offer stable political or cultural prin-
ciples. Renewed prosperity and Muslim prestige rested largely on contacts
with the expansive economy further east; it was from the 'Abbas! domains
that cultural fashions were set in Spain.

But these cultural fashions were so much more attractive than what the
Spaniards had been used to, that they were readily adopted by all the popula-
tion. The leading Christian elements in the Muslim-ruled area tended to share
Islamicate culture, learning Arabic more than Latin. But they did not lose
the consciousness of belonging to a wider Christendom. Their church con-
tinued its ties to Rome.

While in the dominions of the caliphs the area under Muslim rule was slowly
being enlarged, in the amirate the area brought under Muslim control by the
first conquests, which had included almost all Spain and much of southern
Gaul, was steadily eroded away. The Frankish dynasty of northern Gaul, and

310 THE CIVILIZATION OF THE HIGH CALIPHATE

most notably Charlemagne, easily drove the Muslims out of Gaul and made
contact with the remaining pockets of unconquered Spaniards along the
mountainous northern border of the peninsula, an area rather answering to
lands south of the Caspian which the caliphal state was then incorporating.
These Christians in turn, under petty kings in several little states, soon ad-
vanced at the expense of the Muslim power, whose main centres were in the
more fertile and populous south. Before long, Spain was divided between a
prosperous Muslim-ruled south (centred on Cordova and the Guadelquivir
basin) in regular contact with the east Mediterranean Muslims, and a smaller
zone of Christian-ruled kingdoms in the north. Between the two zones were
the Muslim marches, northward from Saragossa in the northeast, from Toledo
in the centre, and from Merida in the west, normally (but often insecurely)
under Muslim control but organized on a war footing; and between the Mus-
lims and the Christian states war was normal.

In the Maghrib itself, the Berbers were left to themselves with very little
Arab immigration to provide alternative leadership to that of the dominant
tribal groupings. But Islam had yet posed a new basis for the recurrent tribal
combinations. Other religious allegiances, except the Jewish, soon disappeared.
(Or the practices associated with them were overlain and reduced to subordin-
ate elements in the religious life.) In the far west of the Maghrib (Morocco),
another refugee from the 'Abbasids, Idris b. 'Abd-Allah, an 'Alid who had
taken part in a Shi'i rising (786) at Mecca after al-Mahdi's death, persuaded a
number of tribes to accept his lead as a descendant of Muhammad. He him-
self lived only long enough to provide a tomb which became a shrine for all
the area (Mawla'i Idris), but his son, Idris II, established a dynasty which
retained, from its urban centres, and in several rival branches, the allegiance
of a diminishing area through the High Caliphal Period. Idris II (in 808)
founded (or refounded) the inland city of Fas (Fez), which became a centre of
international commerce and culture. More important than any political role,
the Idrisid presence became the starting point for extensive missionary
work among the population, especially by immigrant 'Alids and their
progeny, who could count on the tribesmen's respect on account of their
descent. So, the Berbers learned to identify themselves with traditions
stemming from a wider world.

In the more central and eastern Maghrib, where fertile land did not extend
so far inland from the sea, the pressure of the caliphal power and of Arab
settlers had been stronger and Berber resistance from the time of Hisham
accepted the leadership of Khariji theorists, some of whom had come from
the Iraq or Arabia. An 'Abbasi governor was able to establish himself at
Qayrawan (Kairouan), dominating the eastern Maghrib (by 761), but under
'Abd-al-Rahman Ibn-Rustam (by 778) Ibadi Khariji leadership was able to
establish a state in the central Maghrib (now Algeria) centred at Tahart
(Tiaret). This state flourished under a series of imams, successors to, but not
necessarily descendants of, Ibn-Rustam, who were held more or less respon-

312 THE CIVILIZATION OF THE HIGH CALIPHATE

sible to the Muslims that recognized their command. This state was hospitable
to Khariji and non-Khariji Muslim refugees from elsewhere, notably from
'Abbasi rule, and proved prosperous; its merchants took advantage of the
trans-Saharan trade which was increasing along with the Mediterranean trade.

In the eastern Maghrib (now Tunisia and Tripolitania), there was no need
to open up the area to contact with wider currents; from the time of Carthage
it had been a commercial centre. Here, it was the 'Abbasid governors them-
selves who became independent, in the line of Ibrahim b. Aghlab, al-Rashid's
governor who had put down the Kharijis there (800). Al-Rashid had ex-
empted the province from control by the central bureaurcracy and required
only a lump-sum payment from its revenue; under al-Ma'mun, the Aghlabids
made their own policies with little reference to the caliph. It was probably as
much a growing commercial prosperity, in which the shipping from Muslim
lands predominated and gave Muslims an advantage, as the by now tradi-
tional expectations of Arabs and Berbers for new lands to loot and conquer,
that led the Aghlabids during the ninth century to occupy Sicily and several
parts of southern Italy taking over from the Byzantines; Sicily remained
mostly under Muslim rule for about two centuries.

Despite the close ties of the eastern Maghrib to the east, the lands of the
whole Berber-associated region, both the Maghrib itself and Spain, main-
tained close contact among themselves. Ultimately, they developed in com-
mon a special pattern of Islamicate life. When the 'ulama.' subsequently crys-
tallized their Shar'l law into rival schools, the Maghrib and Spain were the
chief areas that accepted Maliki law when most of Islamdom accepted Hanaf i
or Shafi'i. They even developed a special form of the Arabic alphabet, differing
slightly from the form used elsewhere. The dynasties that grew strongest in
the west Mediterranean were likely to extend throughout that area, and with
them fashions of art and education. 6

' Writers concerned only with European history often set off Islam in the western
Mediterranean countries, as 'Western Islam', from Islam in the eastern Mediterranean
and beyond, which they lump together as 'Eastern Islam'. In terms of European history,
where the Latin-German West is set off from the Greek-Slavic East, this twofold division
makes some sense. As applied to Islamicate history, to call everything east of Tunisia
'eastern' is absurd. Even Mecca is west of centre. The usage may derive a specious war-
rant from the fact that the modern Arabs can usefully be divided in this way into western
and eastern. But in Islamic history as a whole, such a division has yielded only a number
of confusions.

It is true that Spanish and Maghribi Islamicate culture had a history of its own, dis-
tinct from that in the central lands from Nile to Oxus. This warrants some special
terminology. But it does not, except when one is explicitly writing from a Maghribt view-
point, justify lumping all the rest into a single complementary category, 'eastern'. The
lands to the north, to the south, and to the east of the central area also had their distinct
traditions. One can — at some risk — distinguish, i.a., between far western Islamdom and
central Islamdom. But to make a twofold division in which everything except the west
Mediterranean lands appears as 'eastern' is to reinforce the Europocentric illusion that
Spanish, Sicilian, and Maghribi Islamdom formed something like half the whole. It has
helped falsify, for instance, the history of Islamicate science and philosophy by facilitating
the notion that these shifted from 'east' to 'west' and then died out with Spanish Islam;

THE ABSOLUTISM IN FLOWER 313

Alternatives to the caliphate

In the main body of Islamdom, most lands remained longer in close associa-
tion with the caliphal state. Even here, however, there were movements that
sought to dispense with the central absolutism; but it was only locally — where
they were most remote from the Fertile Crescent and Iran — that they had any
great success.

Under the early 'Abbasi regime, as under most agrarian regimes, there were
revolts from time to time by peasants who resisted providing the raw material
for all the splendour of the cities. The possibility of such revolts did set a cer-
tain limit to what the city classes could hope to squeeze from the peasants.
But the revolts played no major role in the life of the time. Occasionally, like-
wise, factional struggles — for instance, warfare in Syria which took the form
of the old fighting between Qays and Qahtan Arabs — led to local resistance
against the central power. From time to time Khariji bands still gave trouble
and desperate 'Alids appealed to the loyalty of the Shi'ah against a dynasty
that was betraying Islamic idealism. Eventually the Kharijis and Shi'is were
able to build permanent footholds in several parts of Arabia and of the Magh-
rib. But none of these quarrels proved a major threat to the absolutism, with-
in the region from Nile to Oxus, before the end of the ninth century.

The most serious rebellions there were a series of efforts by Iranians, more
or less Islamized or frankly Mazdean, to shake off at least locally the Arab
dominance. The earliest of these, already in the time of al-Mansur, seems to
have set the keynote for a number of less active movements, by setting out to
avenge Abu-Muslim, the chief general of the 'Abbasi revolution in Khurasan,
making him posthumously a defender of the Iranian tradition against Arab-
ism. For a time most of Khurasan fell into the hands of al-Muqanna', the
'veiled prophet', an associate of Abu-Muslim, finally obliterated in 780.
Some of these movements seem to have combined elements of Shi'i millen-
nialism with sectarian forms of Mazdeism. 7

It was under the 'Abbasids that the last independent princes of the lands
south of the Caspian were finally reduced; but many of them continued to
wield their hereditary powers under Muslim control. It was in those lands that

a notion itself resulting from the habit of lending significance to Islamicate thought only
to the degree that it was translated into Western languages, in which process Spain (and
Sicily) naturally were prominent. A more banal example : the Encyclopaedia of Islam
article on armies has a primary section referring chiefly to the central areas (called the
'Muslim east'), and noting exceptions for the 'Muslim west' (Maghrib and Spain), but not
for the south or the north, where conditions were just as divergent; nor, of course, for the
truly eastern lands. Then there is a long special section on the 'west', but none on north,
south, or east. This is one of the many cases where Muslim history has been dealt with
from a purely European viewpoint — as a parameter, if not a function, of Occidental his-
tory. In general, the categories 'West and East', like 'Orient' or 'Asia', are relevant at
best only to European history and must be avoided in other fields.

7 G. H. Sadighi, Mouvements religieux iraniens (Paris, 1938), gives a not altogether
balanced but yet very useful survey of these movements.

314 THE CIVILIZATION OF THE HIGH CALIPHATE

the old Sasanian tradition was kept most vividly alive. The last and most
nearly successful of the Iranian movements, however, was that of Babak, who
established his independence in Azerbaijan early in al-Ma'mun's reign. It
seems that Babak inherited the following of those who claimed the spirit of
Abu-Muslim, and based his movement on special religious teachings, probably
chiliastic, with both Mazdean and Shi'i Muslim elements. But he gained the
support of some from the respectable landed classes. He had allies in the sub-
Caspian provinces, won some support from the Byzantines, who treated his
state as an independent power, and held out for two decades (to 837). For
centuries after, there were small groups in the Iranian countryside who traced
their chiliastic hopes to Babak.

In Arabia, as in the more distant Maghrib, the caliphal absolutism was
avoided locally while the prosperity it stood for helped maintain relatively
sophisticated social conditions. But any independence there was achieved on
a far more limited scale. At Nazwa in the 'Uman, in the later eighth century
the Ibadi (or moderate) Kharijis made themselves independent, maintaining
ties with Tahart. Only in the later ninth century was a Zaydi (or moderate)
Shi'i state at Sa'dah in the Yemen able to establish itself — at about the same
time as a more transitory small Zaydi state arose in the sub-Caspian moun-
tains. Like the lands of the Maghrib on the routes to the Sudan and western
Europe, the more fertile of the lands of Arabia seem to have flourished largely
through the trade routes which linked the caliphal lands with the coastlands
of the Southern Seas.

By the time of al-Rashid's son al-Ma'mun, in the early ninth century, the
most active parts of the population of most of the caliphal state were be-
coming Muslim, and the caliphate had been reaffirmed as an absolute mon-
archy answering to the expectations of the mass of its population. No alter-
native had proved viable within the primary region of caliphal power, the
historic lands from Nile to Oxus. Correspondingly, the culture of that region
was coming to be carried on in Arabic, and all the major dialogues of the high
culture of the following centuries were well launched in their new Arabic
forms: the courtly tradition, centred on a literary adab and on Hellenistic
learning, and the Shar'i tradition of the Piety-minded 'ulama' among the
bourgeois. It was within the Islamic tradition, likewise, that the more active
forms of religious concern and personal piety were developing. The social con-
cern and factional disputes of the Piety-minded were yielding to a broad
range of religious activity answering to the broader spectrum of the popula-
tion that were now Muslims.

The Shar'i Islamic Vision,
c 7 50-9 4- 5

The first task of the Piety-minded after the 'Abbasi triumph — still more than
before it, when they still had some political hopes — was to elaborate a Shari'ah
religious law that would be admitted as binding on all Muslims. Many of the
'ulama' were judges under the 'Abbasids, like Abu-Yusuf. Most theorists, per-
haps, were not. But at this period they did their work with the practical prob-
lems of actual government in mind. Their thought set off from the situation
as they knew it from Marwani and early 'Abbasi times and they did not lose
sight of the masses, their disciples. At the same time, much of the impelling
emotional force was provided by their need to bring their whole lives into ever
tighter accord with the divine will as found in Qur'an and hadith.

The Islamic aspirations: universalistic, populistic spirituality

In all the confessional religious traditions there was a tendency to wish to
reform all social patterns in accordance with demands set up by religion, with-
out adulteration from earlier, secular ideals. In China, to take one extreme,
any such wishes developing in the more markedly confessional traditions
failed almost totally. The occasional triumphs of Buddhist or Taoist religionists
against Confucian philosophy, in the centuries just before Muhammad, did
not prevent the Confucian elite from regaining a dominating position in
Chinese social life — -though not without introducing into Confucianism itself a
cosmic orientation coloured by the confessional religious outlooks. In Europe,
on the contrary, the Christians did succeed in eliminating the social power of
the Platonist and Stoic philosophers, though their triumph was not completely
unchallenged till the sixth century, shortly before Muhammad's time.

But even in Europe, the Christian triumph was limited. The same Justinian
who closed the Platonist Academy of Athens promulgated the greatest of
Roman law codes. This has perpetuated throughout Christian Europe the
legal thought of the great Pagan lawyers of Greco-Roman antiquity in sub-
stantially its original form despite its Christianized tone. And, though the
Academy was closed, Plato's works as well as the whole dramatic, epic, and
historical corpus of Pagan Greece (or Rome) continued to inform the higher
education of Greeks and of Christian Europeans generally. The head of the

315

316 THE CIVILIZATION OF THE HIGH CALIPHATE

state was the old Roman imperator, unmistakable even when converted to a
new faith.

In the Sasanian lands, on the contrary, Christian bishops built their own
codes for their flocks, codes which can be called Christian; but even here the
challenging standard of the Sermon on the Mount relegated to more or less
stopgap status all replacements of the Jewish law that were less exacting
than the Sermon itself. Even in the lands from Nile to Oxus, the monks repre-
sented the true conscience of the Christian world, protesting against a life con-
formed to a semi-Pagan society; though they might accept the existing condi-
tions for Christians generally, for themselves they wished to build quite new
and specifically religious institutions. In monastic communities, the laws of
daily life, social order, and (to cite the Benedictine example) even labour were
alike created afresh. The godly life was their explicit purpose. But they formed
a minority community apart, dependent on the ungodly society of the world.
The Christian spirit did not favour building all society on a monastic model.

Among the Jews and the Mazdeans, on the contrary, there was an explicit
effort to build a code of personal and social life which should spring in every
detail from the received principles of religion. All adherents should equally be
subject to the all-embracing religious requirements, while being married and
carrying on the ordinary work of the world. The Piety-minded Muslim attitude
to the demands of religion upon life turned out, when fully formed, to resemble
the corresponding attitudes among the Jews and Mazdeans more than it did
anything among Christians. But it differed from each of these in significant
ways.

The Mazdeans developed a detailed code of personal and social behaviour
based on concepts of ritual purity and impurity and on a sanctified social
stratification, in which the priests of Ahura Mazdah reserved to themselves a
prominent position. The role of the king under the blessing of Ahura Mazdah
was to maintain the divinely ordained social order, on the basis of an elect
aristocracy in an imperial nation. The Sasanian empire thus represented a
serious attempt to order all society in terms of religious insight. It rejected the
pattern of Buddhism and Christianity, of restricting such a total religious de-
mand to monastic communities.

Rabbinical Judaism, as it is reflected in the Talmudic writings of the Iraq
and Palestine, likewise created for its elect people a general code of personal
and social life on the basis of explicit religious legislation, and likewise re-
jected the monastic alternative. But unlike Mazdeism, after the fall of Jeru-
salem it no longer effectively had any real priesthood, but in its place had a
class of learned men, rabbis. The rabbis interpreted the law but otherwise
were not differentiated, in point of relationship to the divinity, from any other
Jews. In the same spirit, there should be no other aristocratic class. All Jews
were essentially equal.

By the time of Muhammad, Judaism was very important numerically be-
tween Nile and Oxus, particularly in the towns of the Iraq, where its most

THE SHAR I ISLAMIC VISION 317

widely recognized chiefs came to reside. It is clear that a significant part of
the population that accepted Islam in its formative centuries was composed
of Jews, whose narrative traditions, called Isrd'Uiyydt, dominated the popular
legendry of early Islam. It would appear that much of the spirit that formed
Muslim expectations of what a religion should be was inspired by Jewish
example. Political accidents from the time of the Riddah wars on through the
rise of the Piety-minded opposition to Marwani rule had assured a key position
to socially-minded Muslims: more particularly to those Muslims who saw
Islam as imposing political responsibility, and yet saw this responsibility as
shared on an equal basis among the Muslims — not being left to the rulers or to
any official organized body. Such Piety-minded Muslims, by the time of
'Abbasi rule, had been conceded at least a veto power over what was to be
considered legitimately Islamic. This strategic position was confirmed in the
following centuries, as the wider population became Muslim, by those cir-
cumstances which had already been at work to develop a populistic spirit in the
Irano-Semitic monotheisms: a spirit hitherto most fully respresented in
Judaism.

Like the Jews and Mazdeans, those Muslims recognized as most authori-
tative, when the Islamic tradition was fully formed, came to take very serious-
ly the aspiration of religion to form all ordinary life in its own mould. Most
forms of Muslim piety rejected monasticism as withdrawal from the social
obligations laid upon men by religion itself. Particularly among some sections
of the Piety-minded, the supposed example of Muhammad was to be followed
throughout ordinary life, in matters small and great, from brushing the teeth
to deciding life and death. The Islamic Shari'ah (adj. Shar'i), or sacred law,
was at least as universal in scope as the Jewish law, the Halakha. In a remark-
able number of details, the practice finally espoused by Muslims in their
Shari'ah law was parallel to that inculcated by the rabbis. Like the Jewish,
moreover, the Muslim ideal allowed of no aristocratic class order such as the
Mazdean. All Muslims were to be equal.

But (after the establishment of the Piety-minded ideals) the Muslims, un-
like the Jews, did not regard their own community as a unique and (in prin-
ciple) hereditary body selected out from a world left otherwise without direct
divine guidance. The Muslim community was thought of as one among many
divinely guided communities such as the Jewish or the Christian, all (at their
origin) equally blessed. Thus far, Islam took explicitly the form that various
Christian and Jewish bodies had implicitly been assuming under the confes-
sional empires — an autonomous social organism with its own law for its own
members. The difference between Islam and the other communities was that
Islam was first to rule over and then to supersede all others. Islam was to
bring the true and uncorrupted divine guidance to all mankind, creating a
world-wide society in which the true revelation would be the everyday norm
of all the nations. It must not merely guide an autonomous community like
the Jewish; it must guide the practical policies of a cosmopolitan world.

3l8 THE CIVILIZATION OF THE HIGH CALIPHATE

Thus Islam was formed in terms of the expectations which had been fostered
by the great monotheistic religious traditions, in particular as they had been
understood in the Fertile Crescent and Iran. It chose the social attitude of
Judaism and Mazdeism in particular, in the egalitarian form of Judaism. But
it introduced its own conception of universality, which gave it a radically
fresh approach. We subsume the most central of these aspirations, as they
were embodied in concrete norms of life in the High Caliphal Period, under
the name of 'Shari'ah-mindedness'. For it was in the development and
exposition of the Shari'ah law, and in insisting on its central position in the
culture, that these aspirations most effectively expressed themselves. '

The Shar'l image of the pristine Medina

From Abu-Hanifah (d. 767) at Kufah, and Malik (d. 795) at Medina, who
developed a criticism of Muslim law as it prevailed under the Marwanids,
through al-Shafi'i (d. 820) and his successors, who established a comprehensive
legal theory based upon that criticism, the practical growth of Shari'ah law
was complex. But pervading this growth was a persistent common tendency
which received ever more adequate expression as the Shari'ah was developed.
Though each particular step in the formation of the Shari'ah had its immedi-
ate rationale, there were inevitably many potential alternatives. That the
major choices prevailed as they did was surely due to their enabling Muslims
to come closer to fulfilling the overall ideals of the Shari'ah-minded. These
ideals they did not present in the abstract manner required by the historian,
who measures them against the corresponding ideals of other eras. We must
state in our own modern terms, and against the background of the ages that
had preceded, what it was that those early Muslims were taking for granted;
what it was that they were acting upon without articulating. But we may
hope to come to a formulation which, while they would not have made it,
they would not have repudiated once they understood it. Once we have seen
the overall tendencies that went into making the Shari'ah of High Caliphal
times, we can study more clearly the detailed process by which it was realized.
Central to the Qur'anic challenge and a keynote of Islamic faith, distin-
guishing it to a lesser or greater degree from that of most of the other religious
traditions, was an emphasis on direct and universal human responsibility
before God. The Shari'ah-minded carried to their furthest implications the
populistic and moralistic tendencies already important among their predeces-
sors. For them, this meant first that every person, as such, with no exceptions,
was summoned in his own person to obey the commands of God: there could
be no intermediary, no group responsibility, no evasion of any sort from direct
confrontation with the divine will; and, moreover, that a person was sum-

1 Basic problems in the role of Islamic faith in the historical life of the Muslim peoples
are studied by W. Montgomery Watt, Islam and the Integration of Society (London,
i960).

THE SHAR I ISLAMIC VISION

319

moned to nothing else; anything not in accordance with the divine commands
was frivolous if not worse. The demands of the faith were personal and total —
or at least so the pious saw them, who tried to work out a social order on such
a basis.

The Masters of Fiqh

SYRIAN LEGISTS

Influenced by Syrian
and Umayyad
governmental
practices

al-Awza'i
(d- 774)

ME DIN AN LEGISTS

Claimed closer memory
of Prophet's Sunnah, and
less taint of innovations

Malik b. Anas

(7 J 5-795)
'Maliki school'

IRAQI LEGISTS

Used analogy but also
preferred equity; influenced
by Iraqi and 'Abbasid
governmental practices

Abu-Hanifah

(d. 767)
-'Hanifi school'

Abu-Yusuf
(d. 798)

al-Shaybani

(d. 805)

al-Shafi'i
(d. 820)
'Shafi'i school'
Rigorous care to verify
hadiths, especially of
Prophet's Sunnah; use of
analogy

I

Ibn-Hanbal

(d.'855)
'Hanbali school'
Emphasis upon using
carefully chosen
hadiths; preference
for 'a weak hadith
over a strong analogy'

Dawud b. Khalaf

(d. 883)
'Zahiri school'

Emphasis upon and restriction
to literalist use of Qur'an
and hadiths of the
Prophet's Sunnah

There could then be no question of a church, ministering God's grace to
humans, nor of priests whose ritual acts mediated between a group of wor-
shippers and God. It was symbolically correct that in public worship the

320 THE CIVILIZATION OF THE HIGH CALIPHATE

leader, the imam (who might be any one of the faithful) performed the same
acts as anyone else, only standing in front of the others, who made their
gestures in time with his. With the rejection of Arabism as the basis of Islam,
there could not even be a chosen people. Though Muslims were set off from
others, it was only in that they had chosen personally to obey God, a duty
incumbent upon everyone else as well.

But so far as Islam was to govern social life as a whole, such attitudes had
consequences beyond the sphere of how the cult itself was to be organized.
The Shari'ah law could recognize no hereditary social class structure, for all
Muslims must be on the same footing before God, the only legitimate distinc-
tion being a person's degree of piety. Nor could it recognize a territorial dis-
tinction, such as we Moderns are used to, of nationality, whereby a political
boundary line determines an individual's rights and duties: his only rights
and duties were those laid down by God, and these were the same everywhere.
Indeed, it could not even recognize a state, properly speaking an organization
responsible as an organization for seeing to public concerns. For any public
concern was either a matter of fulfilling God's will, or it was illegitimate
frivolity; but if it was a matter of fulfilling God's will it was a matter incum-
bent upon every individual, at least in principle. There could be no corporate
limited liability through which the individual could evade his duty. In short,
no man, no institution, no human structure of any sort could legitimately be
vested with any responsibility which could relieve the individual Muslim of
his direct and all-embracing responsibility before God.

And yet Islam could not be a community of hermits with no common
institutions: on the contrary, part of God's command was precisely that per-
sons should live in community, should worship together, should marry and
bring up children, and even should see to it that right and justice prevailed in
the earth among all human beings. The substance of human beings' obligation
to God was to be God's vicegerents on earth, ordering all things aright. One
of the basic duties of any Muslim, which had been recognized from the time
of the Piety-minded opposition, was to 'command the right and forbid the
wrong' — which meant not only mutual exhortation to right action among the
faithful, but universal responsibility for public good order. The Shari'ah law
could not ignore social duties, even if it refused to legitimize any formal
organization for carrying them out. Some sort of pattern, then, must be found,
on the basis of which humans could, in principle, without abandoning their
individual responsibility, actively order their community and indeed the
whole world according to the divine norms. The pattern found was both
ingenious and daring.

The heirs of the Piety-minded opposition naturally found the answer in the
community which Muhammad had established in Medina, and which his close
associates had maintained with Medina as their capital. But now it was no
longer, as in Marwani times, a matter of providing a political norm, primarily,
with social patterns supplementary. Rather, it was the social pattern that was

THE SHAR I ISLAMIC VISION 321

to form the essential, and any political implications were derivative. The
rejection of bid' ah, innovation, was erected into a system of law.

It was especially among the legists of Medina, naturally enough, that this
ideal was worked out in its purest form. The legists of each province at first
identified their own local traditions with original Islam. Al-Awza'i (d. 774),
the foremost legist of Syria, regarded the original tradition as unbroken up till
744, the beginning of the third fitnah, and frankly decided questions on the
basis of Marwani practice. Those of the opposition could not be at ease in such
an approach. Yet the legists of Kufah, disciples of Abu-Hanifah and his peers,
continued to identify Kufan practice with original Islam. At Medina several
legists, of whom the most judicious, or at least the most influential, was Malik
b. Anas {d. 795), in insisting that it was their local tradition that perpetuated
original Islam, introduced the new concept most clearly. Malik worked on the
explicit assumption that the ways recognized by the pious elders of the Ansar
of the Prophet's city were uncorrupted either by the indifference of the tardily
converted Umayyads or by the tribal ways of the garrison towns; hence the
ways of Medina went back not just to early Muslims but to Muhammad him-
self. Malik composed a comprehensive collection of rules, the Muwatta', de-
signed to preserve this old Medinese lore, which became the foundation docu-
ment for a whole school of legists — active even in his day as far as Spain.

A somewhat idealized picture of Medina emerged for this purpose. The
Shar'i theorists interpreted that community in terms of the religious attitudes
of which I have just given a modern and oversimplified formulation; and, so
interpreted, the Medina community served well as pattern. In the pristine
Medina there had in fact been neither church and priest, nor state and law, in
any ordinary sense. There was instead Muhammad, who demonstrated em-
pirically in his own actions the commands of God; for it must be assumed that
he would not act except in accordance with them. Accordingly, the first
principle which the pristine Medina provided to the later theorists was a new
basis for law. This was neither legislation by a human assembly nor deduction
from a set of human principles but, rather, empirical observation of individual
actions which God had approved. For Malik, what mattered was still the
typical action, that which Medina had acknowledged; later even rare or al-
most unobserved actions could become normative. These might be either the
actions of Muhammad himself or those of others to which he did not object;
the essential was not the man who had acted but the act that was approved. 2

1 Too often, modern writers tend to assume current Western patterns as normal and
to account for Muslim deviations from them on the basis of the accidents of Muhammad's
life and Arab circumstances of his time. The present discussion may seem to encourage
such analyses. Of course, Islam did not 'unite church and state' at all (as often stated) but
rather — never having had such an institution as a church — it did not invent or adopt for
itself the distinction between the two (though it did adopt some other distinctions which
the Occident in turn did not make). But even this different evolution had its own his-
torical reasons : while the shape of Islam in Medina reflected local conditions there where
no state or church had been, yet one cannot say that the later Muslims failed to legitimize
a distinct state (or church) because Muhammad had lived in Mecca and Medina'. There

322 THE CIVILIZATION OF THE HIGH CALIPHATE

A second principle was also provided by the pattern of Medina: the basis
for organization within the community — that is, for the distribution of tasks.
For duties that not everyone could or need fulfill, Muhammad had appointed
individuals who were personally responsible for carrying them out; and if they
did so, others were not given the burden. Accordingly, theorists divided all
duties into fard'ayn, which everyone was obligated to whether others per-
formed the duty or not, such as worship or keeping one's contracts; and fard
ki/dyah, which were inherently incumbent upon all in the community, but
which could be left aside by others if some one person performed them satis-
factorily. Thus if some saw to it that the mosque was kept in repair or the
weights used in the market were true, others were relieved of the responsi-
bility; but they were relieved only provisionally and on the assumption that
the one responsible fulfilled his task. Nor could the man responsible be content
to leave a matter to subordinates, for the duty was not one of an office, but a
personal one. A story was told of how 'Umar, the caliph, personally delivered
food to a needy widow who had been overlooked by his lieutenants; for he
knew he could not assert, as defence on Judgment Day, that his responsibility
was fulfilled by appointing the best possible state functionary. On the basis
of this principle there need be no constitution and no separate public law;
what might be called public duties were to be handled legally on almost the
same basis as private ones.

A third principle formed the basis of the relation of the community to non-
adherents. It was the mission of the community to bring God's true ways into
all the world; hence the rule of the Muslim community should be extended
over all infidels. As in Medina, however, these latter (if they were adherents
of a former revealed religion, which had merely been corrupted, such as Jews
and Christians) were allowed to continue in their own religious allegiance, as
dhimmis, 'protected subjects', so long as their worship was not too blatantly
inconsistent with the public recognition of God's unity, and so long as they
submitted to Muslim control of general affairs. Thus only those who had per-
sonally undertaken the obligations of Islam were expected to live the Muslim
way, but only they were allowed to bear responsibility for society at large.

Implications of the Medina ideal for High Caliphal times

All the religious communities have been accused of a certain dissociation be-
tween the explicit grounds adduced for their tenets and any analyzable his-
torical actuality. Thus the main body of later Roman law, which made no
religious claims, adduced as its basis precedents going back to a jus gentium,
common practice among Mediterranean peoples, interpreted in the light of
Stoic philosophical principles; and in fact, historically, this was substantially

was no Shari'ah at Medina either. Rather, because of Muhammad's experience Islam was
able to serve as a good vehicle for the Irano-Semitic communal tendencies, which took
advantage of the Medina example to enforce an ideal valid for more general reasons.

THE SHAR I ISLAMIC VISION 323

its formation. But so soon as religion enters in, this frankness tends to dis-
appear. We are confronted with dogmas — statements it is held immoral to
disbelieve; hence explanations which may be rational enough in themselves
must be put in such a form that the dogmas are kept intact. Thus in the case
of Roman ritual, when once it was ascribed to Numa, the Romans no longer
dared explain its importance rationally in terms of social solidarity and of
dramatizing the moral and the sacred. They had to explain it in terms of
Numa's ancestral wisdom — which came to the same thing in a roundabout
way.

The Shar'i doctrines of social order undeviatingly maintained such a dissoci-
ation, in the form they finally took, perhaps in part because of the apparent
rationality and clarity of the few dogmas involved. We must for the moment
go behind the way in which the early Muslims formulated the ideal, so as to
analyze in our own way the Medina ideal as it applied to High Caliphal society.
That society differed doubly from that of the pristine Medina. Under the Mar-
wanids, to begin with, the Muslims had been a ruling minority in a vast rich
empire, rather than a small compact rural community. Later yet, the bulk of
the population of that empire was becoming Muslim. How could the old social
pattern be applied to the new situations? Let us review in this perspective
what had been happening already in Marwani times among the Piety-minded.

The community in Medina presented three distinctive features which all
the legal principles traceable to Medina presupposed. Above all, it presented
an ideal orientation: Muhammad's whole aim was to build a godly life, and
the community responded to him in this aim. Second, it presented a personal
relationship within the group — each person knew everyone else and, being
treated as a responsible servant of God, did not behave, and was not behaved
to, in the impersonal manner of an official functionary. Third, it presented a
cultural homogeneity within the group; and it was this homogeneity — one of
common Arab custom as modified by the group allegiance to Muhammad —
that made possible the other two features: the immediacy of the ideal orienta-
tion, as well as the immediacy of the personal relations among the faithful.
The problem posed, then, was to reproduce in the caliphal society, spread out
over thousands of miles, the ideal-oriented, personalized, cultural homogeneity
that had prevailed in pristine Medina. Such questions as that of authority, on
which the various factions of the Piety-minded differed, were given their
urgency by the terms of this problem. When we consider in its light the various
solutions proposed to such questions, each solution is seen to have reasonable
grounds, whatever the bizarreness of the terms in which it is formulated.

Islam had to prove capable of forming the whole basis of a society. To this
end, it was necessary for Muslims to find what, in the pristine Medina, would
form the basis for an Islamic society as such; and what, in the circumstances
first of Marwani and then of 'Abbas! times, would answer to the relevant
elements of the pristine Medina, so that the Islamic society could be repro-
duced.

324 THE CIVILIZATION OF THE HIGH CALIPHATE

It was this cultural homogeneity that was referred to in the demand for
sunnah practice as a standard of Islamic life. In the Muslim empire the homo-
geneity of the pristine Medina was represented by continuing elements in the
actual life. The Piety-minded took advantage of these in their effort to recap-
ture the ideal. Most generally, the cultural homogeneity of Medina was re-
presented at first by the common traditions of the Arab ruling class, which
retained the old Arab sunnah, customary practice, as modified by Muham-
mad's reforms. What they had in common was the old and indeed largely pre-
Islamic Arab ways; what they had taken over from the life of the several con-
quered provinces was not included except so far as it had been generalized by
the Marwani administration. In recognition of the importance of this Muslim
Arab homogeneity, the Piety-minded had established the principle of adher-
ence to tradition (at first, to local tradition), which later finally crystallized
as the doctrine of ijmd' , that whatever had been accepted generally by the
community was to be regarded as sanctioned by God. It was on this basis
that the use of circumcision (male and female) was imposed, for instance,
though it had no special sanction beyond Arab custom, and the more careful
legists could not regard it as fully binding. The prohibition against bid'ah,
innovation in practice, corresponded to the assertion of ijma'; its tendency
likewise was to preserve homogeneity. This prohibition was aimed against
every sort of deviation, inherently immoral or not, though the legists some-
times relented in favour of obviously helpful innovations.

More particularly, the legists could depend on the existence of inherited
knowledge about the pristine Medina itself. Though the immediate example
was no longer present, ways could be found of recapturing it for the wider
community. This was the intention of the legist Malik in recording, for all to
read, the practice in Medina in Marwani times, as it was known to the oldest
and best-informed inhabitants. But others were not sure that the later Medina
could be relied on to afford the ideal standard desired. Despite Medinese pride,
the practice of every Muslim centre seemed to go back just as clearly to the
first Muslims. What came increasingly to substitute for the immediate ex-
periencing of the pristine Medina was the collecting of eyewitness reports,
hadith, about thousands of details of its life, transmitted by men and women
whose concern for preserving the record of that life could be depended on.

This method had the added advantage (though it was not usually noted as
such) that mingled with these reports was the collective wisdom of two or
three generations of Muslims gained in a variety of experiences throughout
the Muslim empire. 3 This wider experience was expressed both in sifting and
in inventing hadith reports. On both counts the hadith corpus in its classical
form must be ascribed as much to Marwani times and later as to Muhammad's

1 For the process whereby Marwani legal practice was transformed into the ideal
Shari'ah, see Joseph Schacht, The Origins of Muhamtnadan Jurisprudence (Oxford
University Press, 1950), especially part III, chap. I. Perhaps he gives too much credit
to al-Shafi'i.

THE SHAR I ISLAMIC VISION 325

own generation. Nevertheless, a unity of spirit was maintained through insis-
ting on showing that the reporter's central concern was with the life and ideals
of Muhammad's community — that, as it was put, the reporter was sound in
faith and the report was consistent with the Qur'an. Forgery did little harm
if the common spirit was adhered to, and might even foster homogeneity in
points otherwise left in doubt. 4 (The legists themselves, to be sure, rarely
forged a hadith report; they made use of what they found available.)

The 'ilm knowledge embodied in this hadith came to be considered by the
Piety-minded as the highest knowledge and in fact the only really legitimate
knowledge; learning it was incumbent upon every Muslim so far as he was
able. It was no longer possible, as it had been in Medina, for a man to take
these things as 'common knowledge', merely asking a respected neighbour if
for some reason he needed to know a particular point; the personal contact of
Medina life was missing. But, as in the case of cultural homogeneity, a certain
approximation to this condition of personal interrelationship was possible. In
Marwani times, not all Muslims knew each other personally; but the prominent
Muslims (whether in politics, poetry, or any other line) in each of the relatively
few communities in which the Muslim class lived could be and were known, as
personalities, to informed Muslims elsewhere. It was, therefore, still possible
for Muslims at large to have an opinion of the piety and the dependability of
individuals active as transmitters of hadith reports or as representatives of
'ilm knowledge generally. By travelling from centre to centre, a concerned
young man could learn personally from all the more active men of his day; in
'Abbasi times such travelling became very common. Hence the search for 'ilm
on the part of the pious youth could conceivably be carried out on the basis of
a personal relationship between the older and younger generations in the com-
munity as a whole, without institutionalization or theoretical abstraction.

In Marwani times, then, the garrison-town society, with its Arabism, pro-
vided an effective context in which to develop the principles of an Islamism
which was to reform it. Personal interrelationships such as the Medina com-
munity had known could be maintained, within the circle of concerned pious
persons at least, by travel among Muslim centres, and the solidarity of the
society was further reinforced by many details of the Islamic code as that was
built up. The cultural homogeneity needed to make these details effective was
valiantly striven for through the cultivation of hallowed practice, sunnah,
and the opposition to bid'ah, innovation. As to the ideal orientation of the
Medina community, that was provided with enthusiasm — even if only on a
minority basis — by the Piety-minded themselves, strengthened by the ever-
present Qur'an. In this way the model of Medina became relevant to later
Muslim society.

* Compare on the rise of the hadith reports especially Ignaz Goldziher, Muhammedani-
sche Studien, vol. 2 (Halle, 1890). Joseph Schacht has added precisions to our under-
standing of the legal hadith in particular, in The Origins of Muhammadan Jurisprudence,
cited above, especially part II, 'The Growth of Legal Traditions'. Various other subse-
quent studies have not yet made Goldziher out of date.

326 THE CIVILIZATION OF THE HIGH CALIPHATE

Under 'Abbasi conditions, when Muslims were numbered in the millions,
such concerns and principles had to be reduced to a complexly technical sys-
tem. It was no longer true that every important Muslim could be known in a
significant way to all the well-informed, nor could cultural homogeneity based
on essentially Arab family traditions be maintained. Accordingly, the system
adumbrated in Marwani times had to be institutionalized in such a way as to
retain as much as possible of its original virtue. Where the face-to-face re-
lations at Medina had been replaced by relations based on general repute
within a ruling class, these now had to be formalized into a relatively im-
personal relationship of colleagues in a common specialty. The pious specialists
became themselves a distinct class, as 'ulama' scholars; and principles that
had earlier included in their scope the wider Muslim community in an informal
way, now often applied to this religious class alone. Moreover, their applica-
tion no longer depended on informal attitudes but followed a strictly codified
set of rules. The custom, the local ijma', of the garrison towns had to be re-
placed altogether — in principle — by the use of reports about early Medina;
reports which were circulated as text and were often explicitly contrasted to
local tradition. (In fact, of course, such reports did embody what was essential
of the Arab garrison-town tradition.) But all this further transformation had
already been prepared for in the less formal conditions of Marwani times. The
adaptation maintained close continuity with the ways of the older Piety-
minded groups: for they had made the essential leap in substituting Islam for
Arabism as the norm of the community.

Shar'i theory and al-Shafi'i: hadith from the Prophet

Both the Shi'i and the Jama'i-Sunni 'ulama' scholars worked out systems of
Shari'ah law in response to the developing need. Here we shall describe,
rather summarily, the Shari'ah as worked out by the Sunni 'ulama'. The
radical Shi'is, at least, differed on the theory they used — they accorded also
to hadith from the imams the dignity Sunnis ascribed to hadith from the
Prophet alone, for instance, and rejected as transmitters of it those who they
felt had betrayed Islam. Yet in their concrete results, the Shi'i forms of the
Shari'ah were not very distinctive except on points specially relevant to Shi'i
piety. Dealing with the same questions, all the systems found comparable or
even identical answers. The Shi'i systems seem to have received their rigorous
discipline after al-Shafi'i had disciplined the Sunni systems, and probably
their principles, like those of other Sunni systems, were set forth in response
to those of al-Shafi'i. It was the Jama'i-Sunni systems that took actual effect,
for the most part, so far as the Shari'ah was then embodied at all in civil
practice.

The hadith had been, at first, any reports of primitive Muslim practice
(sunnah). Hadith reports were cited on the authority of diverse prominent
early Muslims; when they were authenticated with a chain of guarantors, an

THE SHAR'l ISLAMIC VISION 327

isndd, this chain led back only so far as the figure regarded as authoritative
for the point in question — who might or might not have been an immediate
associate of the Prophet. Under al-Mahdi (775-785) and al-Rashid (785-809),
it had become increasingly accepted that hadith reports traced back to
Muhammad himself had precedence over reports traced merely to his associ-
ates or to the first caliphs; and such reports were being, in fact, increasingly
traced to Muhammad. (Thus an early position of the Shi'is was being accepted,
that decisions of the first caliphs, as such, were not binding.)

It was the legist Muhammad b. Idris al-Shafi'i (d. 820) who took the most
decisive steps. Born in Syria of Hashimid family, he grew up an impoverished
orphan in Mecca, then studied as a young man with the legist Malik in Medina.
As an official in the Yemen, he joined in a moderate Shi'i rebellion and was
taken prisoner under the caliph al-Rashid; thereupon he came into contact
with the legists of Baghdad. It was evidently safer for him, however, to launch
his own ideas in Egypt, where he did most of his subsequent teaching.

Taking advantage of the growing number of reports claiming to go back to
Muhammad, he abandoned definitively reliance on the local tradition of any
one centre, even Medina, in favour of relying on reports about the Prophet.
This was to carry the Medina ideal to its logical conclusion and at the same
time to liberate legal thinking from traditional limitations, giving full play to
the ideals developed among the Piety-minded — in whose circles it was, that
reports about the Prophet had been proliferating. But this consequential, if
somewhat risky, position was reinforced by al-Shafi'i's rigorous legal think-
ing. In detailed legal argument he much refined Abu-Hanifah's pioneering
legal thinking. At the same time, he attempted to create a logically self-
contained legal whole which should derive every detail of law from incon-
trovertible Islamic principles; and he was successful (given his assumptions,
particularly about the validity of those reports that were accepted among the
Piety-minded).

Al-Shafi'i proposed to show, in the introduction to his main law book,
Kitdb al-Umm, that it was possible to derive the whole Shari'ah from the
Qur'an (and hence from the immediate human-divine confrontation which
had taken place historically and which the Muslim community was founded
on). This was an unpromising task; as we have seen, the Qur'an is, on the
surface, very inappropriate as a legal text. If it does make rules, it makes
them for the occasions of a developing community, and commonly softens in
the next phrase what it has fulminated just before. Yet al-Shafi'i started with
the Qur'an and showed, by an incisive critique of its language (in which he
brought out all the diverse ways it called for interpretation), that that Book
presupposed Muhammad not only as its deliverer but as its interpreter.

This gave Muhammad and his sunnah practice a theological status more
exact and far more determinative than they had had before: Muhammad was
to have more than the generalized prestige of having been the vehicle of re-
velation and the first commander of the Ummah community. He was also the

328 THE CIVILIZATION OF THE HIGH CALIPHATE

divinely certified exemplar, whose practice itself had a revelatory status: it
was through his personal words and acts, and only his, that the commands of
the Qur'an could be legitimately interpreted. In comparison, the practice of
even the closest of his early associates had at best merely a presumptive value
in the absence of direct evidence as to what the Prophet himself had done in a
given contingency. But Muhammad's sunnah practice, in turn, was to be
known not by tradition, in which the individual had to follow blindly what he
could not test, but by the text of hadith reports, which the individual could
learn personally just as he learned the Qur'an, and which he could {if quali-
fied) personally evaluate. It was hard for a pious Muslim to resist the idea
that, if the Shari'ah could in fact be built on such principles, this must be
accepted as the ideal to strive for by all suitable means.

The doctrine al-Shafi'i espoused was the most rigorous form of the new
preference for Prophetic hadith: that reports from the Prophet not only had
precedence but were alone authoritative. Moreover, the Prophetic origin of a
report must not be left implicit nor be merely asserted without proof; it must
be shown by the isnad of the particular report, the list of persons through
whom it had reached the current speaker, which must be unbroken all the
way from Muhammad. Al-Shafi'i hoped that such a test would eliminate in-
novations introduced after the corruption of the Medina community. For in
principle the practice (sunnah) of the Prophet himself, so attested, must be
unquestionable. In practice, the results were more subtle.

The isnad, to which little attention had been paid anyway in the early use
of hadith reports, had rarely if ever originally led back explicitly to Muham-
mad, even if the report happened to be about Muhammad. At best one of his
associates might be mentioned. When an explicit isnad leading to Muhammad
came to be required, this was often supplied to reports already current — •
whether they were explicitly about Muhammad, or merely offered an opinion
which must now be supposed to go back to him. Presumably, the more honest
analyzed the probabilities of the case and supplied a full isnad to any given
report on the basis of what they knew of its provenance and of the sources
from which their own teachers had been in a position to draw. On the other
hand, some of the pious had little hesitancy at simply inventing isnads — and
in fact the hadith reports themselves as well — in a good cause ; for they assumed
(quite explicitly) that whatever was true and of value for Muhammad's com-
munity must have been said by Muhammad, as an agent of Providence,
whether it was actually recalled by anyone or not — or even whether it had
actually passed from his lips. (We have hadith reports ascribed to Muhammad
making Muhammad assert just this!) Accordingly, a considerable body of
hadith was soon available with the required isnad documentation going back
to Muhammad.

But this body of hadith reports was not a mere conglomeration of anyone's
forgeries. For in the generations following al-Shafi'i, this potential source of
invalidation in his system was remedied: the wholesale acceptance of newly

THE SHAR I ISLAMIC VISION 329

invented reports was limited by 'isnad criticism'. By analyzing the isnad
chains of guarantors, the 'ulama' scholars were in fact able to assure a selection
of hadith with characteristics appropriate to their needs. Both in the substance
of the report and in the selection of the chain of transmitters to father it on,
the attitudes of the person who had put the hadith into circulation were mani-
fested. Potential transmitters were known; men with reputations for improper
doctrine could be ruled out.

At this stage, isnad criticism amounted to selecting those reports that were
circulated by persons who respected the same sorts of authorities as did the
critic, and who showed this respect by using them in the isnad. This did not
make for historical accuracy, as to what Muhammad himself had said; but it
did make for an effective religious homogeneity. The method was of course a
bit unsure; what was more dependable was that, for the future, it would be
relatively easy to sort out hadith reports which had not yet been accepted by
the community at the time when the standards for the isnad documentation
became formulated: roughly about the time of al-Shafi'i himself. To be sure,
even in this respect isnad-forging was not impossible. The determination of
the validity of hadith reports on the basis of the isnad never fully overcame
the inherent difficulties in the method. Yet in at least a rough way it served
effectively to formulate, and then to preserve, a self-consistent body of doc-
trine and practice recognizable by and acceptable to the Jama'i-Sunni 'ulama'
as a body. Moreover, as demanded by the individualist spirit of the Piety-
minded, this recognition and acceptance took place without recourse to either
hierarchical authority or councils or any other human instrument that would
come between the individual human critic and his conscience.

Al-SMfi'i's system of ustil al-fiqh

Al-Shafi'i's system had to make some allowance for traditional custom and
even for the individual legist's judgment. But the scope of these was rigor-
ously delimited; they were made to derive from his Qur'anic system. Once
this body of hadith reports with their isnads going back to the Prophet had
been developed as expressing the community insight — or rather, the insight
of those held to represent most closely the 'Qur'an and the sunnah of Muham-
mad' — it was possible seriously to reduce the arbitrary element remaining in
particular legal decisions.

This element was labelled ra'y, personal judgment. It could not be elimin-
ated completely from the cheoretical justification of the legal positions that
came to be held; even hadith reports could not be expected to cover all pos-
sible contingencies. But a law dependent on the initiative of every new lawyer
was obviously intolerable. Various principles had been suggested for limiting
the arbitrariness of ra'y. Some who distrusted the reliability of hadith reports
preferred to appeal to 'aql, reasoning, when the reports were not decisive;
hoping that sound reasoning would lead to universal results, and so avoid

330 THE CIVILIZATION OF THE HIGH CALIPHATE

arbitrariness. Some suggested that a decision must be in accord with equity
or else with the public interest. For those who were devoted to hadith, how-
ever, the most effective way proved to be to subordinate an individual's judg-
ment to hadith, in that unprecedented decisions should be made at least on
the analogy of decisions found in hadith or Qur'an. This process of drawing an
analogy was called qiyds, and al-Shafi'i adopted it as a major adjunct to
the use of hadith, for it multiplied greatly the effective force of individual
reports.

One more principle was needed to cap the system. The custom of a given
centre had been recognized as authoritative when it could claim consensus
among the weighty men there — ijma'. Such ijma', consensus, had played a
basic role in the legal thinking of the Piety-minded. In addition to being
conservatively practical, it represented, in maximally concrete form, the ideal
requirement of cultural homogeneity, without which it seemed impossible to
create a society in which each man was immediately responsible before God
for everything. Now al-Shafi'i insisted that the only ijma' that carried author-
ity was that of the whole Ummah community of Muhammad: on the
grounds, in effect, that so wide a consensus must have originated with at
least the tacit approval of Muhammad himself. But even this he was able to
justify more concretely by discovering a hadith report assuring that Muham-
mad's community would never be agreed on an error. Hence what they all
agreed on must be sound; if an explicit report was wanting on a given point
to illustrate Muhammad's sunnah practice, the agreement of the Muslims
would bear witness equally well to what that sunnah must have been. (Cer-
tainly it was the surest guide available for reconstructing it.) In principle,
the Muslim community in this sense meant the whole body of the faithful, or
of the weighty among them; but for more technical purposes it eventually
came to mean the 'ulama' as they expressed themselves in their recorded fat-
was, that is, decisions in points of law or conscience. If all recognized 'ulama'
were on record as accepting a position, it was to be considered binding.

Al-Shafi'i's method was intensely factualistic, allowing almost no leeway for
private fancy. He based his method on quite concrete events: the coming of
certain words to certain people under certain conditions; and the meaning of
these events must depend on the exact meaning of those words to those people
under those conditions. (He noted, incidentally, how important it was to have
an exhaustive knowledge of the nuances of the Mudari Arabic grammar of
Muhammad's time.) Thus he gave full recognition to the intensely historical
nature of the revelation that had launched Islam; though not, except uncon-
sciously, to the continuingly historical nature of the reception of that revela-
tion in the community committed to it.

The method was also legally effective. By rejecting the authority of prece-
dent, of custom, he gave the pious more leeway in building a law to their
taste. But by the precision of his method, he reintroduced, once the new norms
were accepted, more determinacy and predictability into the law than had

THE SHAE'l ISLAMIC VISION 331

existed on the old basis. Above all, he went far toward ruling out all arbitrary
decision, ra'y, by the judge purely on the basis of a personal sense of justice:
the judge must show some foundation for his decision in what the Piety-
minded had accepted as sound; and he had to prove that foundation by rigor-
ous criteria, linguistic and logical.

Al-Shafi'i established a basis for a law that should be at once founded on
ideals and also (most important in any legal system) uniform and predictable.
He did so at a price. In their very lack of system, the earlier legists were
capable of a practical realism which, after al-Shafi'i, could sometimes appear
only backhandedly and in despite of the ideal. More important, al-Shafi'i's
system was necessarily, if unwittingly, founded on a pious fiction.

Al-Shafi'i himself clearly believed he was founding it on the man Muham-
mad, as he lived at Medina. In fact, since he accepted the existing body of
hadith as representing Muhammad, he was founding it on the figure of Muham-
mad, as this had been built up in pious circles; and therefore on the community
as a whole so far as it was committed to Muhammad's vision. Clearly some, at
least, of the later collectors of hadith were aware that it was the figure that
mattered; that the figure was a legitimate and necessary enlargement of the
man. This is implied in one of the hadith reports that justified accepting
dubiously historical reports: it makes Muhammad say that, in judging the
validity of reports about him, the Muslims were to accept any report that was
consistent with the Qur'an: that is, of course, consistent with what meant
most to Muhammad — at least, as this was understood by the Piety-minded.
This was to throw the authority back to that section of the Ummah on whose
good judgment, in fact, al-Shafi'i was relying. From a historian's viewpoint,
so to include within the act of revelation itself the tradition that sprang from
it may seem only reasonable, and not misrepresentation at all. But to a less
sophisticated mind, such a procedure, if detected, would look like direct
fraud. To al-Shafi'i (could he have recognized what he was doing without
first imbibing a modern historical viewpoint), to have to admit that he was
ascribing later judgments to Muhammad himself would have invalidated his
whole system.

Al-Shafi'i was an eminently good legal thinker and his suggestions had,
moreover, an obvious appeal. His requirement with regard to hadith was soon
quite generally accepted. But with this requirement came a great stress on
criticism of the isnad documentation, which meant criticism of the various
reporters through whom the report had been transmitted. Everyone knew
that many hadith reports said to go back to the Prophet were forged. In time,
the transmitters were less generally known; offhand ways of sorting reports
became insufficient. The 'ulama' naturally looked for surer ways of eliminating
false ones. A special discipline grew up devoted to testing the isnads of all
reports in circulation. Critics of the isnad were to trace the recorded character
of each transmitter and reject those whose virtue, memory, or judgment — or
doctrine — were suspect. If there were only sound transmitters in the isnad,

332 THE CIVILIZATION OF THE HIGH CALIPHATE

this should prove that the report actually went back as far as it claimed and
had not been invented by the unscrupulous or mixed up by the forgetful.

There resulted eventually a vast body of specialist literature, dealing both
with personalities and with standards to be maintained in judging them and
their reports. The personal contact of acquaintances in the primitive Medina
was thus replaced by the learned analysis of the hadith literature and its
authors. The evaluation of the worthiness of each man as a source of authority
and knowledge was no longer a matter of common assent among neighbours,
nor even a matter of notoriety among a group of limited communities as in
Marwani times. It was henceforth a technical task, to fulfill which an elaborate
special study was developed with its own rules, the 'Urn al-rijal, 'knowledge
of the men'. This 'ilm al-rijai was used both in sorting out hadith reports
and in deciding whose judgment was to be included in settling points of
ijma'.

The corpus of hadith continued to grow and the added reports were even
utilized in law. But pious Muslims learned (in effect) to discriminate between
the earlier, 'sound', hadith and later reports which might be useful but were
not so dependable. Six major collections of reports were accepted eventually
among Jama'i-Sunnis as canonical (even though some legal positions were
still based on hadith not there included). Two of these, the collections of
Bukhari and of Muslim, came to be revered as especially holy because it was
felt the reports they contained were sifted by the most careful tests of genuine-
ness; the other four allowed 'weaker' reports the benefit of the doubt so as
to afford a somewhat wider basis for legal decision. (Shi'is had their own
collections.)

The hadith corpus, however, was never merely a source for law. It had its
own autonomous character even in point of doctrine. Inconvenient individual
hadith reports sometimes became so well entrenched that they had to be ex-
plained away (thus some of the 'Alid-loyalist reports about 'All could not be
denied by Jama'i-Sunnis and had to be emasculated by exegesis) . The overall
weight of hadith sometimes went directly counter to the positions ultimately
taken in Shari'ah law. The hadith corpus became a reservoir of pious opinion
that sometimes could be used even against the Shari'ah-minded. 5

Legal fiqh

After al-Shafi'i's time, some legists, especially among the devotees of hadith —
representing a form of piety to which this line of thought was congenial —
tried to push al-Shafi'i's system even further. As the body of hadith reports

5 Even the hadith corpus, as distinguished from legal use of hadith, is not properly
regarded as 'tradition' (save in the sense that, once launched, it formed a cultural
tradition of its own). R. Brunschvig, in particular, has shown how greatly the positions
upheld in the hadith corpus could differ from received attitudes; see his 'Considerations
sociologiques sur le droit musulman ancien', Studia Islamica, 3 (1955), 61-73.

THE SHAR'i ISLAMIC VISION 333

became gradually larger, it became convenient to base directly on hadith de-
cisions which had originally been accepted on other grounds. Even qiyas,
analogy, was eventually rejected by some uncritical extremists in favour of
hadith pure and simple. Ijma' was suspect among some — perhaps as being
based on tradition rather than on explicit text — and they allowed that au-
thoritative status only to the consensus of the first generation of Muham-
mad's associates, as shown in hadith reports.

But by and large, al-Shafi'i's system became normative. A systematic
science of law and ethics and cult was recognized, the fiqh, jurisprudence,
through which the Shari'ah law was determined in detail. Fiqh necessarily
dealt with every case of conscience, not only in ritual but in private inter-
personal relations. In this latter realm, such questions were discussed as when
a Muslim should or should not tell the explicit truth. (Christian moralists have
condemned Muslims for allowing lying on occasion; but the Muslims were dis-
cussing cases where every society has in fact condoned or even required lying,
and were attempting to introduce restraint: notably cases assimilable to war-
fare and cases in the area of courteous speech, where evasion can take the
form of misstatement without real deception.) But in the most intimate ques-
tions, such as this, the decision of qadis was rarely invoked. Moreover, other
sources of ethical judgment (sometimes more demanding) intervened to guide
the social conformists or the philosophically emancipated or even the specially
pious. Hence the greater part of the efforts of men of fiqh went into settling
ritual or the less intimate interpersonal activities: what may be called ex-
pressly legal fiqh.

The fiqh was based on four 'roots' (tisiil al-fiqh): the Qur'an, the hadith,
ijma', and qiyas; among them these four 'sources' were supposed to be exhaus-
tive. Each of the 'sources' had been referred to in Marwani times; but they
were now linked systematically and comprehensively. Even those who opposed
the system accepted its terms of debate. Hanafis defended the use of what
others called ra'y, personal judgment, by their early masters, on the basis of
appeal to a supplementary principle, 'preferability', similar to supplementary
principles that even al-Shafi'i had to introduce to guide qiyas. 6

The legal precedents and rules provided by Qur'an and sunnah, even thus
elaborated, naturally did not provide a full law, however. They were the
sources of fiqh, but fiqh had to operate on them by an elaborate system of
rules before they could become a fully developed and processed Shari'ah, a

6 Later Jama'i-Sunnis interpreted the relations among the four madhhabs which they
accepted, in such a way as to make their common existence intelligible and their con-
tinuity from Muhammad authoritative despite their differences. They presented each one
atemporally as the work of a single founder with a given attitude to fixed 'sources' of law;
and pointed out that each tolerated the others as Muslims, even though as erring ones.
Such a picture has been taken up by some modern writers, but it is, of course, anachro-
nistic. The early legists had no need to assert ra'y, for instance, as one of the formal
usul al-fiqh, but at most only as common sense against a few hadith-minded extremists.
Originally, all the schools of legists, and not merely the Hanafis (as later) were accused
by the Hadith folk of using ra'y, which was practically synonymous with fiqh, 'inquiry.'

334 THE CIVILIZATION OF THE HIGH CALIPHATE

legal system. 7 First of all must be decided the relevance and degree of applica-
bility of any given rule or precedent. For instance, some prescriptions followed
in Muhammad's time had been temporary and superseded by more permanent
ones, as turning toward Jerusalem in worship had been superseded by turning
toward the Ka'bah in Mecca. As much as possible should be known of the
occasions for the revelation of particular Qur'an verses and of the chronology
of events in Muhammad's life of which hadith reports might be transmitted,
so as to know what had been superseded by what. Likewise, it must be deter-
mined of any given report whether it presented a duty incumbent on Muham-
mad alone (such as proclaiming a new divine message) or relevant only to a
particular occasion (such as the behaviour of certain Muslims when 'A'ishah
was accused of infidelity), or whether it had a more general bearing. The most
important distinction of this sort that had to be made in fiqh was as to the
degree of legal weight borne by a particular rule: whether a given practice was
obligatory or merely recommended; whether, if it were obligatory, it was to be
enforced by penalties in the courts or left only to the rewards and penalties of
the other world; and indeed (to complete the picture), whether a given act
might be subject neither to an obligatory rule, pro or con, nor even to a recom-
mendation, but be indifferent and so permitted at discretion. 8

But at least as important was a set of considerations only barely adum-
brated in sacred texts. However complete the body of legal rules might be, it
could not be used with precision without an equally complete body of legal
definitions. What is a sale, what is a gift? What is property? To a degree the
answer to such questions, also, could be cast in the form of hadith; but for the
most part it must depend on careful reasoning. It was in this realm that the
masters of fiqh showed their finesse. It was especially in this sort of question
that the Roman law (and doubtless other law less well known to us) seems to
have been a source of the Muslim fiqh. Roman concepts may have entered
especially by way of the law applied by Christian bishops; and the Jewish
example was naturally also strong. 9

' It will be seen that there is no single appropriate rendering of Shari'ah into English.
'Law' by itself answers to only one aspect of it unless one has clearly in mind the 'Law' of
Moses. Moreover, a great deal of actual Muslim law was never taken up in the Shari'ah
and even remained contrary to it, but was applied in Muslim courts. 'Sacred law' scarcely
extends the meaning of 'law' without the same special reference to Moses, though it does
remind one that not all law was Shari'ah. 'Canon law' carries some analogies but also
some wrong ecclesiastical associations. A word like 'code' might suggest the scope of the
Shari'ah, but the Shari'ah was never codified in the strict sense.

* Frederic Peltrier pointed out that the collection of hadith by Bukhari, in contrast to
that by the earlier Malik, allows very expressly for distinguishing when an act, however
uncommendable, has full legal effects and when it is legally void: such distinctions being
essential in practical law courts, though unimportant when the 'ulama' were still con-
cerned primarily with cases in conscience. See his he Livre des ventes du Mouwatfd de
Malik ben Anas, traduction avec eclaircissements (Algiers, 1911), Preface.

9 S. G. Vesey-Fitzgerald, 'The Alleged Debt of Islamic to Roman Law', Law
Quarterly Review, 67 (1951), 81-102, grants only the possibility of some influence.
Elsewhere he points out Jewish influence: 'Nature and Sources of the Shari'a', Law in
the Middle East, vol. 1, ed. M. Khadduri and H. Liebesny (Washington, 1955), pp. 85-112.

THE SHAR I ISLAMIC VISION 335

Thus fiqh had become a highly technical process, based on debate through-
out the Muslim community. The traditions of fiqh which had grown up in
each main provincial capital now were crystallized into schools, consciously
disputing with each other about both the method and the detailed rules of
fiqh. They came to be called after the most revered master of each. Abu-
Hanifah, the master of AM-Yusuf, was the great imam (here a generalized
term for teacher of 'ilm knowledge) in the Iraq; the fiqh which remained
faithful to the Iraqi tradition was called 'Hanafi'. Al-Awza'i was the great
imam in Syria. Those who clung to the example of Malik b. Anas, in the Hijaz
and elsewhere, were called 'Malikis'.

Each school of fiqh was called a madhhab, a 'chosen way'. As the fiqh be-
came more technical, new madhhab schools arose, tied to an ideological
position rather than to a geographical area. Al-Shafi'i himself founded a new
madhhab school which he finally went to Egypt to teach, which spread widely.
Another madhhab school was associated with Ahmad b. Hanbal, whom we
will meet as the proponent of the populistic Hadith-folk piety; he compiled an
important collection of hadith and his followers tried to use hadith (and
Qur'an) as exclusively as possible. Still later, Da'ud al-Zahiri (the 'literalist')
believed that hadith reports were abundant enough to allow him to construct
a complete system of fiqh with no use of qiyas analogy at all; it was he who
insisted that the hadith must be taken literally and no implications intro-
duced by the initiative of the faqih, jurisprudent; critics feel that he too used
inference without admitting it. 10 Schools of Shi'i fiqh were growing up at the
same time, appealing to the privileged 'ilm of the imams. The Shi'i imam
Ja'far al-Sadiq was made the leading authority by both Twelvers and
Isma'ilis among the later Shi'i sects; the various Zaydi Shi'i imams developed
two chief systems of their own. A separate madhhab was worked out among
those KMrijis who were most active in 'Abbasi times (Ibadis in particular,
the group who had controlled much of Arabia during the third fitnah and had
created a state in the Maghrib).

Each Muslim had to choose which madhhab school he would follow unless
he were a great enough scholar to work out his own way (as did the historian
al-Tabari) ; normally, Muslims naturally accepted the madhhab prevalent in
their regions. In effect, most Muslims, except for Shi'is, ultimately became
Hanafis or Shafi'is, while in the Maghrib and Spain the Malikis prevailed. For
whatever matters could be settled privately — personal ritual and ethics, of
course, and matters like inheritance or fulfillment of contract — a person need
only consult a mufti, 'jurisconsultant', of his madhhab. If a matter was
carried to the point of litigation in a court, the qadi appointed by the
governor ruled according to his own madhhab.

10 Something of the subtlety of the Zahiri position appears in R, B. Brunschvig,
'Sur la doctrine du Mahdi Ibn-Tflmart', Arabica, vol. 2 (1955), pp. 137-79, as that
position was taught much later. For fuller treatment, see Ignaz Goldziher, Die Zdhiriten,
ihr Lehrsystem und ihre Geschichte (Leipzig, 1884).

336 THE CIVILIZATION OF THE HIGH CALIPHATE

The spirit of the law: public order, individual rights

Thus was built in 'Abbas! times a solid structure of fiqh, not exactly 'case law'
but law case by case, which could govern the whole field of ethics, cult, and
private and even public law in a manner clearly derived from, and to a con-
siderable degree actually consistent with, the ideal of individual responsibility
before God's will as expressed in the Prophetic mission. In all madhhab
schools, however much they had at first varied in their stress on this or that
element in deriving the law, the spirit of the law was much the same, and the
same spirit pervaded all its branches.

At least within its range, the system was comprehensive and rigorous.
There was as yet, indeed, no attempt actually to reason out all elements of
social law and practice on a Shar'i basis. Once the position of caliph was
recognized by the Shari'ah, large discretion was left to him in practice to
decide many matters of public policy without explicit reference to Shar'i
principles. The Shari'ah covered, in the first place, matters in which the Arab
soldiery had a direct interest, and then also those in which the merchants were
interested — family and commercial law, above all; other matters were globally
covered under the principle that actions not expressly prohibited in the Shari-
'ah were to be tolerated. Such limitations in scope were confirmed in 'Abbasi
times, at least so long as the caliphate remained powerful, by a further shift in
social outlook among the 'ulama' scholars. As the identification of Islam with
old Arab f amilies receded; as the popular Islam of the lower-class urban story-
tellers became numerically more important — that popular Islam which was
largely a retelling of Christian and Jewish tales from a Muslim viewpoint —
the outlook of the pious leadership, Jama'i-Sunni or Shi'i, also changed. Build-
ing for an audience which could hope for little political role (under pre-Modern
agrarianate conditions) save possibly the essentially negative intervention of
street mobs on behalf of one or another ruling faction, the 'ulama' scholars
turned their interests ever more to questions of private life or perhaps of
factional dogma.

Yet even so, this private life was seen largely from the viewpoint of public
order: public order in worship, public order in the market place, or on the
highway or the frontier. Those responsible for enforcing the law were to take
cognizance only of what appeared in public — that is, to the scandal of the
Muslims. Though minute prescriptions were worked out covering the most
private acts, the qadi was given jurisdiction only over what was brought to
his attention without prying (unless the rights of an innocent party were being
infringed). The Shari'ah was above all the norm of the Muslim community as a
community. i ' However (as already in Marwani times), this was a public order

11 There have been some to see in the Shar'i approach to law and community life — as in
much else where Islam does not make the distinctions Europeans have made — a 'primi-
tiveness' which they ascribe to its being developed among Bedouins (which, of course, it
was not) or at least among 'unsophisticated' Arabs. But to be at once comprehensive and
simple is not primitive; indeed, it is just the reverse. That kind of simplicity has histori-

THE SHAR I ISLAMIC VISION 337

in which individual rights, as we shall see, often took precedence over col-
lective interests.

The first concern of the law was, of course, with acts of worship, the public
ceremonies of the regular salat and of the hajj pilgrimage, as well as of other
special occasions. These were regulated down to the minutest detail. But there
was no precise line between public ceremonies and individual, private ritual.
In the public ceremonies the effect depended chiefly on punctilious participa-
tion by every individual, for there was no priest; the imam served chiefly as
model for the rest. The legists required the same precision on the part of
individuals in all their ritual acts, whether they were being performed publicly
in a group or privately with none observing. These acts of worship were the
homage mankind paid to God and even in the most private detail formed, in a
sense, part of a common universal obligation.

The ritual obligations of the individual were therefore minutely described
on the basis of hadith; the disputes among madhhab schools (often virulent
and even violent enough) usually took the overt form of disputes about ritual
details (based on contrasting hadith reports) such as how a person should hold
his hands in one or another stage of performing the salat. In performing divine
worship, however, almost as much weight was put on ceremonial purity as on
precise propriety in execution. First, plain bodily cleanliness was required (and
the manner of washing was described in detail) but, in addition, purely ritual
ablutions were required after a variety of contacts not very tangibly polluting;
some conditions invalidated the salat altogether — thus a menstruating woman
must not perform it. This concern with externals in the formal homage human
beings paid to God, however, was not allowed to become formalism in the
strictest sense, for God was held to judge by the intention, even in matters of
ritual, rather than by the actual event. But one could not justify allowing a
sloppy performance to slip by on the ground that one had intended better, if
by taking pains one could have done better.

Such minute prescriptions were worked out, above all, for the five main
ritual acts to which every Muslim was in principle obligated (sometimes called
the 'five pillars of Islam') : the formal declaration that there is no divinity but
God and Muhammad is God's messenger (called the shahddah), the salat wor-
ship, the zakat (legal alms), the hajj pilgrimage to Mecca, and the fast in the

cally been very hard to attain. For law to be subsumed under ethics, in particular, is not
necessarily the same as primitive undifferentiatedness, if all the main features of law
do exist: distinction between what bears sanctions and what does not, specialist termino-
logy and knowledge, written authorities, emphasis on predictability, and so forth. These
were all present. But the spirit was directed toward trying to find a transcendent norm,
rather than leaving a human decision as absolute precisely qua human decision. Hence
for Shar'i law, precedent holds only so far as it is ethically correct; for Anglo-Saxon law,
precedent holds qua precedent {for the sake of predictability). Nor is a law oriented to
public life reducible to a 'primitive' identification of morality only with the expectations
of a tribe. It may, rather, spring from a healthy respect at once for individual privacy
and for the need to smooth the ways of public intercourse, such as cosmopolitan merchants
are bound to appreciate.

338

THE CIVILIZATION OF THE HIGH CALIPHATE

month Ramadan. But there were also minute treatments, supported by hadtth
reports, on such matters as the potency of this or that form of words as prayer,
or what historical persons are to be cursed (as public enemies) or blessed (as
divinely favoured) and in what form.

The Derivation of a Shar'i Legal Decision
God

Qur'an

'sunnah (practice of Muhammad)
hadith (report) ^» ijma' (agreement)
isnad (reporters).

qiyas (analogy)

[ra'y (private
judgment)]

'ilm
>• (learning)
of the
'ulama.'
(the
learnSd)

ijtihad (inquiry) by a mujtahid

fiqh (jurisprudence) of a faqlh

Shari'ah (the way for the faithful)

fatwa (advisory decision) of a mufti

I
qada,' (court judgment) of a qadi (judge)

For instance: Muhammad used a toothpick after meals (sunnah); there is a
report to that effect (hadith) transmitted through a chain of reporters (isnad) ;
the mujtahid studies this report (ijtihad) and decides that the use is
recommended not only for the Prophet but for the ordinary believer (fiqh) ; it is
therefore to be included in the Shari'ah, the established way of life; and a
mufti may deliver a fatwa to an inquirer, telling him he should do it; but as
it is merely recommended, the qadi will assess no penalty if he does not.

The law of the market place was almost equally a public concern with the
law of the mosque; here especially, good public order was required. But again
it was not so much collective interests as individual justice that was the
chief concern of public order. In the law of contracts, the Shari'ah insisted
more on substance than on form. A contract was not binding unless it involved
some sort of real equivalence in an exchange, for instance. On the whole it
was expected that most human relations, outside of close friendship and family
ties, would tend to take the form of contract relations rather than be deter-
mined in advance by status; yet many provisions attempted to guarantee
those who were weak in one way or another against the strong taking great
advantage of them. Perhaps rather naively, for instance, contracts that called
for taking interest were banned: a position that was becoming stronger in

THE SHAR I ISLAMIC VISION 339

the monotheistic traditions, and that the Muslim legists now tried to make
absolute by interpreting a certain Qur'anic word as covering any money pay-
ment for money.

It was more or less tacitly expected that contracts, deeds, and the like
would be in writing, but the personal guarantee of an honourable man was
required in any case as providing living assurance which sheer paper could
not. Here, as in some other cases, the heritage of the Arab garrison towns can
be seen: though the Qur'an had encouraged the written contract, the public
word of a tribesman was trusted more than a written word which might be
that of a hired scribe and in any case could be verified only by the literate,
who were likely to be the rich. The qadi, judge, of a town maintained lists of
honourable citizens who could serve satisfactorily as witnesses; some of these
were simply professional notaries, but it became an important honour to be
included in the list.

The contract law was thus tied fairly closely to ideals of honour, generosity,
and mutual aid among the faithful, at once congenial to the feeling of the
Arab tribesmen and to the old populistic spirit of monotheism. For more
'businesslike' affairs, some of its explicit provisions were impractical. Certain
useful types of ground rent were ruled out, for instance; and interest, banned
in principle, has always been a major resource for investment. Already by the
end of Marwani times, legists like Abii-Hanifah were working out ways of
carrying out more 'businesslike' proceedings when appropriate, without set-
ting aside the more idealistic provisions of the law as basic norms. One should,
indeed, not make of a loan to someone in need the occasion of profiting from
his distress; but if two businessmen were agreed, then it might be legitimate
for the one who was to profit by the consequences of a present deal with the
other to be bound to share his profit in a complementary future deal. The
'tricks' (hiyal) which the legists worked out to this end were at first, at least,
a means of safeguarding the spirit of the law. In time, they amounted to a
vast treasury of subterfuges whereby practices that could not practically be
disallowed were given legal status; the most pious, naturally, came to frown
on their use.

From a pre-Modern point of view, the Shar'i criminal law was so mild that
most pre-Modern Muslim rulers felt bound to save their subjects from the
results of applying it intact. It required quick procedure — 'the law's delays'
were anathema to the 'ulama' — and safeguarded the accused in ways that in
the Agrarian Age, devoted to the use of torture and of Schrecklichkeit, seemed
dangerously soft on criminals. Penalties ran to public humiliation, corporal
beatings, and monetary compensations. Mutilation of various sorts was very
commonly used in pre-Modern times and was countenanced in the Shari'ah
also in certain cases. Money fines, which bear far more heavily on the poor
than the rich, were not common. A handful of penalties happened to be pre-
scribed in the Qur'an and were given a prescriptive status. Among these was
cutting off the hand of a thief; but the legists limited this, as far as possible, as

340 THE CIVILIZATION OF THE HIGH CALIPHATE

to amount and circumstances of the theft. A great many deviations from the
ideal were, of course, not penalized on this earth at all.

Family law: pressure toward, equality in personal status

Muhammad had been especially concerned with family law and it retained a
major place in the Shari'ah. A proper regulation of each person's civil status
was the very foundation of public order, and that status was not left to the
free play of contract between individuals. Only a limited range of provisions
in the marriage contract was allowed, for instance, or of inheritance or of other
family relations. No one was free to arrange whatever relations suited him
best: these relations were a concern of the community as a whole. In part, of
course, the regulations were a means of protecting the weaker against the
stronger; but in part they were an enforcement of common norms of public
propriety.

In discussing family law, it is convenient to consider the rights of women,
children, and other dependents as against the male head of the family, who
by nature, if law makes no contrary provision, is in the most advantaged
position (being not only, on the average, stronger than a woman but more in-
dependent, since he is free of pregnancy and the immediate care of children).
Family law largely consists of restrictions upon his presumed freedom. In the
context of the Nile-to-Oxus region as a whole, the Islamic family law, even
when it perpetuated rules laid down by Muhammad, naturally meant some-
thing rather different from what it had in Muhammad's time.

The Islamic marital rules had encouraged individual responsibility in
Medina by strengthening the nuclear family. Against the background of a
formerly Christian and Mazdean population, where the nuclear family was al-
ready strong, the rules encouraged that egalitarianism and social mobility,
especially among males, which Islam favoured in other ways also. Wealthy
men, at least in ages of sharp social stratification, have commonly maintained
more than one woman as sex partner; in the Christian Occident for instance,
until fairly recently, it was common for men of quality to have one or more
mistresses in addition to a wife. (Most men of ordinary means have naturally
had to be content with a single mate in all societies, if only because of the
biological sex ratio.) 12 Among the Christians and Mazdeans (as in most socie-

" In older writings, 'monogamy' and 'polygamy' were usually distinguished as the two
contrasting normal forms of marriage — Christians practicing the one and most other
peoples the other. (As 'monogamy' had prestige value, too often writers intending to
justify various non-Christian social systems tried to show they were in fact monogamous,
sometimes even by taking note that polygamy was practiced 'only by a minority' — over-
looking the fact that this is no special merit, for it is usually guaranteed by the laws of
nature.) This conventional dichotomy unfortunately obscures artificially many of the
most important questions. Not only does it. make too sharp a distinction between systems
that differ much less in practice (given the custom of keeping mistresses) ; it cloaks the
points of real difference. Thus the presence of a slave-based 'harem' system, even with
only one wife, may be socially more decisive than the presence of two unsecluded wives.

THE SHAR I ISLAMIC VISION 341

ties), such wealthy men as maintained more than one woman had to accord
special privileges to a primary mate and to her children. Among the Mazdeans,
the secondary mates received some legal protection and their children might
under some circumstances inherit. Among the Christians, the secondary mates
had in principle no rights whatever and their children were stigmatized as
mere bastards (though in practice such 'natural sons', in contrast to the off-
spring of casual unions, sometimes might maintain a high position, just as in
the Medieval Occident the bar sinister sometimes inherited not only good
breeding and wealth, but even lands and sometimes title). The effect of the
Shar'i rules on marriage was to accord to up to four mates absolutely equal
rights, which their children also shared in; the kept mistress or the free-born
concubine disappeared, in effect, from among the ordinary privileged classes.
Moreover, though the male was permitted to take his slave girls to bed (as
normally happened wherever slavery itself was allowed), girls who became
pregnant by him were granted privileges and their children, if recognized, had
rights exactly equal to those of children of regular wives.

Among wealthy circles (and the fashion of the wealthy tends to become the
norm of those lower on the social scale) the tendency of such a regulation was
the exact opposite of that of the Christian exclusivism. Especially in Mediter-
ranean Europe, where women tended to be secluded and domineered in both
communities, the Christian marital pattern and the Muslim could look very
much alike in the ordinary case; yet in special cases the two legal construc-
tions had sufficiently different consequences to set, cumulatively, a different
tone. The Christian system sanctified — and under favourable circumstances
surely fostered — a solidarity of interest in a couple committed to a single
marital union despite the temptations of wealth. The Muslim system sacri-
ficed the primacy of conjugal unity in favour of equality of rights on the part
of all concerned.

There being no privileged mistress of the household, privileges often associ-
ated with her were also missing; and among ordinary families, with a single
wife, this fact was more important. Here the male prerogative, as the one
who was required to support the household, was protected by the Shari'ah,
and the wife was protected primarily by her family. For a woman to divorce a
man would mean to unsettle her husband's economic investment — and to
exact financial support from her own kin. It was not only prejudice against
women's mental capacities but sound agrarianate economics that insisted on
one-sided divorce privileges. According to all madhhab schools, a husband
could divorce a wife almost at will, if he did not fear retaliation from her
family; but a wife who wished to leave her husband had to show good reason.
The chief legal check upon the man in divorce was essentially financial and a
matter of contract between equal parties: it became customary that, of the

And where there are secondary partners, their status may determine family patterns as
decisively as that of the primary partner, who alone tends to be taken into account in
discussing 'monogamy'.

342 THE CIVILIZATION OF THE HIGH CALIPHATE

mahr, bridal gift, which the man settled upon his wife, only a part was paid
at the time of marriage; if he divorced her without special reason, he had to
pay her the rest of the mahr, which might by contract be substantial; if she
obtained a divorce from him, she forfeited what mahr he had paid. (In the
Shafi'i system, the wife might get a legal separation more easily than a divorce.)

But the equality of the wives among themselves (and inheritance by daugh-
ters and widows) carried with it an important financial independence for the
women. Whereas in the closely tied Christian household what the woman had
was the husband's, the Muslim woman's property could not legitimately be
touched by her husband, who, however, had to support her from his own
funds. This potential feminine independence had to be curbed primarily by
keeping marriages, as far as possible, within the extended family — for instance,
by the marriage of first cousins, so that family property would not leave the
family through women marrying out.

A well-born woman thus had a personal status which might allow her to go
into business on her own (as sometimes happened especially in widowhood),
but she lacked the entrenched dignity of a materfamilias, with a marital and
maternal status assured for life. At the same time, the mingling of all classes
among the males, encouraged by Muslim disregard of inherited rank, threw all
levels together on intimate terms; if men and women had mingled freely, this
could have tended to eliminate any remaining aura of exclusive respect for
well-born wives. If the wives might have been content to take the consequen-
ces of a free and relatively contractual position, the men were not willing to
have it so. It early became a mark of a woman of quality that she was secluded
from all men but her own — in private apartments at home, behind a veil if
she walked abroad.

This had been a common custom on the higher levels, it would seem, in
both the Byzantine and Iranian traditions; the egalitarianism of Islam para-
doxically spread it throughout urban life. With all wives on a level, it became
the practice for any woman with pretensions to high quality to take the veil;
the upper classes were followed by the middle classes; in the end only poor
women who had to labour in the open were left unsecluded in the towns (in
the countryside and among the Bedouin, naturally, the custom was not carried
very far). Thus arose a marked segregation within the society, with men (and
singing girls) maintaining one social circle, in which primary political and
social decisions were made, and respectable women maintaining a separate
circle, very much secondary in the public life of society, but ruling each man's
private home life. (The situation somewhat resembled that in classical
Athens.) Under such conditions, the other marital practices of the upper
classes became relatively generalized; while of necessity most men had but
one wife at a time, divorce became relatively common even in lower-class
circles, and sometimes even a poor man might have more than one wife at a
time.

All these ramifications were not explicit in the Shari'ah. The earlier-formed

THE SHAR I ISLAMIC VISION 343

madhhab schools, notably the Hanafis, tended to give the male an extensive
prerogative (though short of that of the old Latin paterfamilias in many ways) ,
which other madhhabs somewhat cut away. But all the madhhabs presup-
posed a considerable social role for women. The Qur'anic injunctions to pro-
priety were stretched, by way of hadith, to cover the fashionable latter-day
seclusion (and special precautions Muhammad had been allowed to take for
his own women were — perhaps somewhat impudently — adopted by ordinary
Muslims who would not think of imitating his privilege as to the number of
permitted wives). But even so, segregation was not made explicit: the veil
was still presented simply in terms of personal modesty, the female apartments
in terms of family privacy. Women were not required to share in the Muslim
public rites, except the hajj (where the veil, as suggesting social distinction,
was not permitted); but they were held for the daily salat and they were
acknowledged as transmitters of hadith and as teachers. The debilitation of
upper-class women which followed upon rigorous segregation was unforeseen.

Children were under ward of their parents or of a legally established guar-
dian. Muhammad had been especially concerned about orphans (having been
one himself) and the Shari'ah reflected this in attempts to avoid illegitimate
wastage of the orphan's property by a guardian. Children could be married off
very young, normally to mates expected to be of their parents' choice; some
legists even allowed a girl to be married without her consent if still under age,
but then allowed her to repudiate the marriage when she came of age. In any
case, a girl could not be required (or allowed) to marry beneath her. But at
maturity, sons, and to a much more limited degree daughters, gained legal
independence of their parents as free Muslims, largely on their own. The
father could not even threaten them very effectively with disinheritance.
There was no primogeniture: one son was not privileged over another; indeed,
though descendants necessarily inherited, they might receive only a fraction
of the estate: under the influence both of the Qur'an and of the tribal-minded
Arab garrison towns, inheritance was obligatorily shared, sometimes even by
cousins. In principle any large accumulations of family estates must therefore
be broken up each generation.

Most wealthy households had other dependent members: slaves (that is,
persons bound to labour at the will of their master, and the rights to whose
labour could be bought and sold) and freedmen. The slave condition was in-
herited, unless a master acknowledged a slave woman's child as his, but Mus-
lims were encouraged to manumit their slaves, and penance for numerous mis-
deeds included freeing slaves, at least those who had turned Muslim. Sources
of new enslavement were limited. A Muslim could not legally be enslaved, nor
a dhimmi non-Muslim living under Muslim rule. Since there was a constant
tendency to free existing slaves, the supply of slaves came largely from be-
yond the Dar al-Islam, either by purchase or through capture in warlike
expeditions. (Slaves came especially from northern Europe, where the export
of slaves was a major trade, eastern Africa, and central Eurasia.) Slaves were

344 THE CIVILIZATION OF THE HIGH CALIPHATE

to be given considerable rights by their masters, but as a master's abuse of his
position entailed few penalties, a slave's life was in fact largely at his master's
mercy. Correspondingly, a slave had fewer responsibilities; legal penalties for
misdeeds were halved in his case. A slave might set himself up in a trade and
purchase his freedom by contract, which the owner could not break once he
had granted it. Freedmen had the same rights as the freeborn, though in
practice they commonly remained associated with the former master even to
the point of inheritance.

Limitations on collective authority

The overall effect of the Shari'ah was to stress the rights of the individual as
such. Equality was the basic principle, above all among free adult males. Cor-
porate bodies, apart from the Muslim Ummah itself, had no standing, and an
individual derived few Shar'i rights from membership in any particular group.
The insistence on individual interests over collective is illustrated in the rules
about the public right-of-way. The Shari'ah ruled that town streets should
have a certain minimum width at least — and when towns were founded,
streets were commonly much wider than that. But it also allowed the in-
dividual householder to build out upon the right-of-way so long as the passage
rights of individual passers-by were not obstructed. This second right was
given precedence over the first rule: that is, the balance between the rights of
one individual and those of another was more carefully maintained than that
between an individual's rights and the common interest, which was not im-
mediately felt in any one case though it was cumulatively felt by all. In this
way the law itself was in part responsible for the Islamicate towns having
rapidly lost the free open circulation their founders usually intended for them
and being broken up into tiny crooked passageways and culs-de-sac.

An individualistic egalitarianism can favour the strong and the bold over
the modest and the honest. The Shari'ah tempered its equality partly through
special provisions to protect the weak, but perhaps still more fundamentally
through a strong bias toward public propriety and uniformity. Its heritage of
respect for the cultural homogeneity of Medina and then of the Marwani
Arabs now became a pressure for all Muslims to conform to a bourgeois pattern
of life, a pattern necessarily adapted to the average man. Shar'i Islam deman-
ded no 'religious athletes' and discouraged any other special callings. The bold
experimenter was required to show, at least externally, the face of mediocre
propriety. The 'ulama' declared Islam to be the religion of moderation, the
natural way of life: every child was held to be born a Muslim till his parents
corrupted him into another faith.

In this way, God's will was to be established, the members of the community
were to be organized to carry it out, and the community's mission to bring
justice to all was to be assured, yet a purely individual responsibility would
be maintained throughout. Indeed, an imam or caliph was still needed to fill a

the shar'1 islamic vision 345

role Muhammad had played at Medina in presiding over the application of
God's will in detailed instances, in particular in assigning to each his task in
the public sphere. Not, of course, that anyone might replace the responsibility
of each individual before God; but someone must fulfill the particular social
obligation, which only one at a time could fulfill, of exercising command in
public matters. But such a figure became less practically crucial. He was re-
quired, on many occasions, more as an authoritative guarantor of the validity
of the agreed-on 'ilm than as personally active intervener in disputes. Even
where his personal interventions and his appointments made a difference, the
Shari'ah and the opinion of the Muslims in support of it could be expected to
keep his appointees within bounds.

The Jama'i-Sunnis now were willing to settle for whomever the bulk of the
community, the jama' ah, found it politic to accept as ruler, hoping that the
weakness of one man or another would be less harmful to the community as a
whole than would the dissensions any other principle must occasion. The
Shi'is were theoretically less casual about the person of the imam. Their imam
must be the authoritative possessor of 'ilm. Those called 'Zaydis', indeed
(rather like the Kharij is) , still held out for an imam of the house of the Prophet
who would be at once a master of 'ilm and a successful rebel-statesman. But
most Shi'is were willing to acknowledge an imam whose actual position was
solely a matter of his possession of authoritative 'ilm, whether he happened to
hold political power currently or not; which meant acknowledging the power
of the de facto caliph in practice, but not granting him even a moral authority
as guarantor of the Shari'ah.

It was a universal problem among the religious communities, to carry the
demands of religion through in the whole of life. The peculiar Islamic formula-
tion of this problem had called for a solution on the basis at once of a radical
equality of the faithful and of an effective governance of the world's affairs.
The principles we have just sketched present an attempted solution of the
problem. But it must be borne in mind that the development of these prin-
ciples was not purely in response to a highly sophisticated intellectual and
spiritual challenge, though it clearly was this in part. In part it was a more
general moral protest against the sophisticated ways of corrupt society, in
part it was an expression of escapism and even of a primitive ignorance of the
nuances of complex social processes. Such ambiguities are present in any
major social endeavour. Yet what is most notable, surely, is the drastic and
far-reaching attempt that was in fact made to remould society from its very
first principles.

The Islamic Ummah under the 'Abb&sis: the two orders of justice

As envisaged by the 'ulama' scholars, the Ummah, the community, was to be
organized as a political body in terms quite as uncompromisingly religious as
those of any other provisions in the Shari'ah law. The 'Abbasids professed to

346 THE CIVILIZATION OF THE HIGH CALIPHATE

acknowledge the Shari'ah as the rules of life which formed the basis for the
Muslim community, the Muslim Ummah. Yet in the face of 'Abbasi arbitrary
rule, the real hopes of the 'ulama' could not be put in effect in the political
order. But, though the Shari'ah was never allowed to mould the whole social
order, it was always given more than lip service. It is not only a theoretical
triumph of the human mind, but has always had a significant practical role to
play; and this role has often been reflected in turn in the theoretical structure
itself.

What the ' Abbasids did concretely was to require that the qadi courts follow
the Shari'ah as it was being developed by the Piety-minded theorists, a move
which they attempted to implement by persuading those theorists themselves
to serve as qadis. On the one hand, this forced the ideal Shari'ah into practical
forms so far as the qadis' jurisdiction extended. At the same time, the move
imposed limits on the qadi courts which stemmed from the Shari'ah itself. It
went far toward crystallizing the role of the Shari'ah in Muslim society as
ruling in practice only certain sectors of it, even while retaining its ultimate
primacy in the whole.

The Shari'ah, originating in the military courts of the garrison towns, had
been military and had had more than a touch of aristocratic colouring: the
dignity of the Arab tribesman was a major touchstone in it. This was especially
true as to its formalistic procedure: everything was done to respect the per-
sonal dignity of the accused — which meant that conviction was difficult un-
less there were honourable eyewitnesses, for otherwise the defendant could
clear himself by an oath, and the chief penalty was his shame. This suited an
egalitarian army very well, though it might not suit the army's commanders.
But it was never aristocratic in the agrarian sense — it was not suited to a
landed gentry, among whom particularistic ancient precedent was at least as
important as personal honour. On the contrary, apart from the army, it lent
itself fairly readily to modification which adapted it to the mercantile classes:
for merchants, also, who had always to defend their inviolability against both
masses and aristocrats, personal dignity was a very serviceable principle. The
rules of procedure were tacitly modified by the use of honourable witnesses as
notaries in advance — and it was precisely in commercial questions that it was
possible to provide for possible future disputes by way of notarization. As to
any substantive rules that the dignity of a military fraternity had introduced
into the Shari'ah which might conflict with mercantile activities, these were
almost from the beginning modified by hiyal rules, whereby the form of awk-
ward prohibitions was observed while its substance was evaded. (These rules
were adopted first in what became the Hanafi legal school and then in the
Shafi'i school; they were not adopted in the third great legal school, theMaliki,
but there their place was taken by the introduction of a series of new categories
which produced the same effect and which were ultimately formalized by a
rule of accepting judicial precedent, not so much needed in the other two
schools.)

THE SHAR I ISLAMIC VISION 347

Acclimated thus to mercantile life, the Shari'ah did preside over all the les-
ser institutions of the city. The market inspector and supervisor of public
morals generally, called the muhtasib, was the officer whom the small trades-
man, the housewife, the craftsman looked to for daily regulation of city inter-
course. He got his law from the qadi judge, and the more developed was the
Shari'ah, the more his activities reflected it (though in routine matters he was
not bound to Shar'i procedure). His force was provided by the caliph or his
governor, whose captain and his men (the shurtah force) were to see that the
muhtasib's orders were complied with. Accordingly, the Shari'ah became the
basic mould in which the social order flowed at its lower urban levels, at least;
the more this was true, the more difficult was it for the highest levels of the
polity altogether to escape its influence.

But the more the qadi courts, in particular, became governed by the Shari'ah
in the strictest Piety-minded sense, the more their limitations forced the rise
of other courts supplementing them. However suitable Shar'i procedure was
for an independent-minded soldiery, it was hardly appropriate to the sort of
army that an absolutist caliphate increasingly was turning to: a standing
body of troops subject to transfer or dismissal and in general personally de-
pendent on the caliph. More important yet, however much it suited the com-
merical activity of a mercantile class, it was less appropriate to other elements
in urban life, and to other aspects thereof. For many purposes, the personal
dignity which Shari'ah procedure protected seemed less important than
effective prosecution of defendants, which it often hampered. Supplementary
courts not tied to Shari'ah procedure (even if they attempted to adhere to
Shar'i substantive standards) were called for, then, by the very nature of the
Shari'ah. But they were called for also by the nature of the caliphal absolutism.

For the spirit of the absolutism remained profoundly contrary to that of the
Shari'ah. By the side of the great monarch stood his executioner, to behead a
man at a word from the caliph. Here was no question of mufti and qadi or of
the sunnah of Muhammad. If the will of God was involved at all it was, as the
courtier might in awe repeat, because the caliph was the shadow of God on
earth, as had been the Sasanian monarch before him. The caliph must main-
tain justice, and against a great man sometimes only the most summary
justice could prevail, lest the man use his resources and influence to raise a
rebellion and perhaps force his own terms on the caliph.

But it was not only against the great who frequented the caliph's court that
the direct caliphal justice prevailed. As essential part of a monarch's function,
the caliph or his vizier was expected to hold regular sessions to redress the
wrongs of whatever petitioners should appear : initially, wrongs committed by
those in authority, in the course of their administration. The jurisdiction of
such a court was, of course, readily extended. Hence alongside the regular
qadi's court was held the ruler's own court for redress of wrongs, mazalim. In
the courts of mazalim, the vizier or the judge who represented him could take
any measures that seemed necessary in equity without regard to the rules of

348 THE CIVILIZATION OF THE HIGH CALIPHATE

Shar'i judicial due process. The Shari'ah was indeed to be followed so far as
possible in substance, but not if proved a hindrance. Thus obscure cases could
be settled by the weight of probability; powerful defendants could not over-
awe by their respectable position; above all, the torture and the severe punish-
ments that all post-Axial societies, at least outside China, thought indispens-
able for good order (at least on the level of the masses) could be applied at
will. While commercial and family questions remained the province of the
qadi, criminal accusations especially were brought to the courts more directly
dependent on the caliph; and more than one sort of court was specialized off
from the overall court of mazalim wrongs to handle such cases.

Hence the very legal order — at the heart of the Shar'i realm — was divided
into two sectors legitimized in mutually exclusive ways. On the local urban
level, legal decisions were recognized as acceptable so far as they answered to
the Shar'i norms, expressing the populistic culture of the market. But on the
level of overall territorial government, legal decisions, in their ultimate sanc-
tions, rested on norms which answered rather to the aristocratic culture of the
court and of an agrarian gentry; and though the city merchants might prefer
to support the court in any possible struggle with an aristocracy, the culture
of the court remained alien to them.

The legitimation of the caliphal state

A landed aristocracy in the mid- Arid Zone, however, now proved too weak to
dominate the society of the vigorously mercantile cities. This fact is almost
symbolically expressed in the remarkable feature of Islamicate history: that
the third natural locus of high culture, the temple, which under the Sasanians
had supported the monarchy and had been closely related to the court culture
and its agrarian background, had been captured by the culture of the market.
The mosque was dominated by the Shari'ah-minded. This meant that the
local, Shar'i legal sector — and all aspects of culture associated with it — alone
was fully legitimized on a basis that the whole society could honour. The
courtly culture remained essentially illegitimate.

In the immediate situation, such a want of legitimation made little dif-
ference, perhaps. In fact, each cultural tradition carried its own commitment
on the part of the groups dedicated to applying it. The general expectation in
a given group has its way in all those routine cases where personal interest
pulls as readily in one direction as another; and such cases establish the pat-
terns of society. Yet these cultural traditions and their dialogues were not
ultimately separable one from the other. Whenever a crisis occurred in the
courtly order, and new ideas were needed to maintain the absolutism under
the particular current conditions, the very right of the absolutism to exist was
put in question by the presence of the more highly sanctioned alternative
tradition of norms.

In the end, though there was some continuity in the absolutist legal tradi-

THE SHARI ISLAMIC VISION 349

tion, it was only as it approximated to and could be justified by Shar'i norms
that this continuity could be depended on. The Shari'ah enforced itself by the
respect that even its violators perforce paid it; the violations never abrogated
the principle, which stood ready to come back in force at the next proper legal
decision. No other legal system could rise above the level of an ad hoc equity
which each individual could apply and develop at discretion, and which accor-
dingly had little binding force in the public expectation. If the very court of
laws in which the vizier passed judgment was not fully legitimate, then if the
next vizier disregarded the precedents of the last, he was in any case no worse
than any other vizier. On this basis, the culture of the court required a legiti-
mation in Shar'i terms if it was to endure past the crises that would shatter
any ad hoc arrangement.

The 'ulama' gradually made a theoretical place for the court. But it was
grudging and tardy, and did not really afford the court true legitimation in its
actuality — in the spirit it necessarily had to express to fulfill the demands of
the absolutist ideal. Even at that, the theory received full elaboration only
after the caliphal state had in practice collapsed.

After the disappointment the Piety-minded opposition suffered in the
'Abbasid triumph, neither Shi'i nor Sunni 'ulama' scholars were inclined to
allow much of a role to the ruling caliph. Even though they recognized his
position, the Sunnis remained as autonomous as the Shi'is; so far as possible,
they vested legal authority in all the possessors of knowledge, 'ilm, literally
the 'ulama'; who were in turn distinguished as more or less sound by the
autonomous working of a process of mutual recognition within any present
generation analogous to the process governing isnad-criticism as to the past.
The 'ulama' were the true 'heirs of the Prophet' and law was to be established
by fatwa issued by a private learned man, rather than by the decree of the
caliph or his agents. But it was only legal judgment that was, as far as pos-
sible, to be retained in the hands of this 'clerical class'; administration had to
be provided for by a 'political class' even at best. Eventually, the Shari'ah law
came to recognize this. The clerical class led the prayer in the mosque, decided
cases of law in the market, and declared the Muslim's duty on the frontier;
the political class — the police, the tax-collectors, and all the various katibs,
secretaries — saw to keeping the mosque in good condition, preserved order in
the market, and commanded the defence of the frontier. It was here that the
caliph was to have his role.

(It is important to note that this division of labour, as it was eventually
conceived in the Shari'ah, is not a distinction between religious and secular
fields of action. In practice, however, the 'clerical class' certainly represented
an Islamic 'religious institution' which was much less total in its operations
than the Shari'ah theoretically might require, and which answered in large
measure to specifically religious institutions in other societies.)

As we have noted, the political class, to operate, stood in need of individual
men of authority — the caliphs; hence the Jama'i-Sunni forms of the Shari'ah,

350 THE CIVILIZATION OF THE HIGH CALIPHATE

which were applied in practice, finally gave to the caliph a considerable ad-
ministrative role. But in limiting this assignment sharply to administration,
the 'ulama' were denying to the caliph — and to any to whom he might dele-
gate his authority — any truly political role: that is, the independent decision
of ultimate policy which had characterized alike Muhammad and the early
caliphs, and which had launched Islam as a power in the world. If the Islamic
vision was to be preserved, it must be at the expense of its dynamic political
motor force.

For such an unpolitical role, Jama'i-Sunnis came to agree that any of
several processes could theoretically serve for the caliph's selection, provided
the community at large acquiesced; and that he need not be the wisest or the
most pious but must simply be sufficiently qualified to carry out an adminis-
trative task: be possessed of his faculties, have sound and effective judgment
— and come from the generally honoured stock of the Meccan Quraysh. Such
minimal criteria were enough to legitimize the way in which the caliph was
chosen in practice — by designation of the predecessor, within the caliphal
family. But one primary ideal requirement was still insisted on. That there
should be only one man in supreme authority within the Dar al-Islam, the
territory within which Muslims ruled and where the Shari'ah was the law of
the land, was yet maintained, for the sake of unity and peace among Muslims.
Hence as provinces fell away from practical control by the caliph at Baghdad
in later 'Abbasi times, they were constrained by Jama'i-Sunnt feeling to
acknowledge at least the caliph's theoretical authority. Finally, recognition
for the caliph came to be a central point in the Sunni Shar'i demands in the
face of increasing political fragmentation of the Ummah community.

Intellectual patterns associated with Shar'i Islam: the study of history

The Shari'ah in itself, of course, was far from sufficient to define a whole
Islamic culture. Both the Shi'i and the Jama'i-Sunni 'ulama' carried their
concern beyond the law and even the ethics embodied in it to wider reaches of
intellectual life. They built especially on the kind of studies that had inter-
ested the Piety-minded in Marwani times — exact and elegant recitation of the
Qur'an, elaboration of Arabic grammar and lexicography, and history not
only of the Prophet but of his community. To these they added the newer
disciplines that had arisen in connection with the fiqh jurisprudence itself,
such as the 'ilm al-rijal. All these studies combined to develop a general sense
of life and of what mattered most in it which, since it centred in the Shari'ah,
I call 'Shari'ah-mindedness'.

Islamicists have often applied the term 'Orthodoxy' to that kind of Islam —
whether Sunni or Shi'i — which accepts the Shari'ah, the all-embracing sacred
ritual law, as fundamental to the religious life. Such a term cannot be applied
in this sense (if in any at all) to the earliest Muslims, for whom the Shari'ah in
its developed form did not yet exist; but after the processes we have been

THE SHAR I ISLAMIC VISION 351

describing, in which the Shari'ah was perfected and exalted, the notion has a
certain relevance. But because the word 'orthodoxy' can be and has been
seriously misleading when applied to the particular approach to Islam here
indicated, I prefer to use explicitly a phrase referring to the Shari'ah and to its
central role in the outlook. Then I can reserve the term 'orthodox' for any
case where a given position may be regarded as established, either officially or
socially — and such a usage will by no means always coincide with 'Shari'ah-
mindedness'.

Yet the common use of 'orthodoxy' for what I call 'Shari'ah-mindedness' —
provided one recalls always that it can have either a Sunni or Shi'i (or a
Khariji) form — does point to a central role Shari'ah-mindedness has won for
itself among Muslims. 13 There came to be, indeed, many groups of Muslims
who in one way or another depreciated the Shari'ah in its literal sense; but
always the majority feeling placed these outside — possibly sometimes above —
the accepted Islamic norms. The supremacy, or at least the crucial importance,
of the Shari'ah has been accepted not only in most Islamic currents, but even
by most sorts of opinion within each current — whether mystic or literalist.
This has been true even when respect for the Shari'ah, while present in a
group, has not been the only or even the most important element in their
religious life. Indeed, even those who rejected its literal application for them-
selves have usually regarded it as binding on most other Muslims. Hence the
Shar'i system has been something of a constant throughout subsequent Muslim
history.

The practical social order of the Muslim Ummah community was central to
the Shari'ah-minded conception of life. But the intellectual labour required to
elucidate this social order and to guarantee its spiritual validity was granted
its due: for instance, the historical scholarship presupposed in the isnad-criti-
cism by which the acceptability of hadith was assured, and the analytical
reasoning necessary in developing a body of hadith-based law. And just as
the legal and social side of the system was held together in a common spirit,
so was its distinctive thinking. In all the fields of intellectual endeavour cul-
tivated by the Shari'ah-minded, a set of intellectual patterns arose which
shared many techniques in common but also, more significantly, bore the im-
press of a common spirit: a spirit populistic and factualistic, with a persistent
sense of the moral importance of historical events.

We can see this spirit especially in the realm of historical inquiry; for here

13 The adjective from 'Shari'ah' is 'Shar'i'. This refers to the Shari'ah law itself, where-
as my term Shari'ah-minded' refers to a whole complex of attitudes characterizing those
Muslims for whom the Shari'ah has had an unrivalled primacy in religion and in life. (The
use of the term 'Orthodoxy' has serious disadvantages. For one thing, it is often identi-
fied with Jama'i-Sunnism, or with a certain sort of theology — whereas for a long time
most of the Shari'ah-minded rejected all logically formulated theology at all. Still worse,
the uncautious reader — and too many scholars — can get trapped into an identification of
other forms of Islam not only as 'unorthodox' but as somehow more or less 'un-Islamic' —
an attitude we have no right to import into the materials.)

352 THE CIVILIZATION OF THE HIGH CALIPHATE

the technique of historical reconstruction by hadith report and supporting
isnad was quite at home. The same method of criticism of authorities as used
for law or dogma could be used in general history — in a form much less intense
than for legal purposes of course. The chain of reporters might be wholly oral
or one link might be formed by a written text; but the isnad treated both cases
alike, for what mattered was the personal continuity of witness that guaran-
teed, from teacher to pupil, even a written text.

The subject matter of history was also often determined by the Shari'ah-
minded world outlook. For such men an understanding of history was neces-
sary if only because the divine revelation had itself been historical — through
prophets sent to given peoples at given times — and the Islamic community,
in which the godly life was to be lived, was a historical community. The his-
torical inquiry of the 'ulama' had begun with two indispensable themes: the
life of Muhammad, the all-important model; and the evaluation of the trans-
mitters of reports about him, which had come to replace the original direct
homogeneity of the Medina community. Their interests soon widened. But
when in 'Abbasi times a great Jama'i-Sunni legist, such as Ibn-Jarir al-Tabari
(d. 923), wrote a universal history, he was still concerned with tracing the suc-
cess and failure of the various communities that had been summoned to fol-
low God's will, and particularly the triumphs and backslidings of the Muslim
community. Moreover, he was, as befitted a Shar'i scholar, concerned above
all with the responsible behaviour of individuals, not with the working of
institutions as such or even, primarily, with the splendour of kings. He pro-
duced a record of the personal decisions of Muslim souls in the series of choices
which had faced the Muslim community.

From Tabaristan, the south coastland of the Caspian Sea, Tabari came to
Baghdad just too late to study with the man he most hoped to see, Ibn-Hanbal
(founder of the populistic madhhab school of fiqh law). He then travelled
widely in the Fertile Crescent and elsewhere seeking religious 'ilm knowledge
before returning to Baghdad to teach as a scholar. His primary interest lay in
building a more perfect system of fiqh; in the course of this, he created the
most substantial of all Qur'an commentaries, in which he collected the chief
interpretations of every discussable verse, and gave his own point of view so
soberly that it came to be very widely accepted. His history was part of the
same overall effort.

Far from embodying sophisticated historical principles, Tabari's historical
method can seem, at first sight, to rule out any interpretive intent at all — to
consist merely in the driest chronicling of data. Tabari rarely speaks in his own
voice, except in jejune frame or transition passages. What he has to say is told
purely by judicious selection, arrangement, and documentation of verbatim
reports which he has received.

As compared with a historian who modifies the wording of the reports he
receives to fit his selection and arrangement of the facts, and who adds his
own deductions and comments, the verbatim method suffers severe limita-

THE SHAE I ISLAMIC VISION 353

tions. Stylistically, the resultant picture tends to seem a bit confused, especi-
ally to the reader who does not already know (as Tabari's audience would) the
main outlines. The same detail may be often repeated, as it recurs in this or
that separate report. In Tabari's description of the death of 'Uthman, for
instance, numerous reports (primarily telling of some detail early in the story)
go on summarily to report the death itself in such a way that the report would
lose vividness and sometimes point if the portion about the death were ex-
cised; hence 'Uthman dies in the narrative many times before the actual
death is recounted in detail. On a more fundamental plane, Tabari cannot
make his own conclusions explicit; he is like the detective who would give in
immaculate detail every piece of evidence which he has found relevant to his
own private conclusions about a case, but in the end would fail to set forth
his reconstruction of it, leaving you to draw your own conclusions from the
evidence he has set in order before you. Unless you have something of the
mind of a detective yourself, you are likely to miss the point.

For a conscientious writer, the verbatim method has its advantages too.
With strict documentation of each report by its isnad, it allows an accuracy of
detail which a conscientious modern historian must envy on occasion. More-
over, if the evidence is misleading, at least the historian has provided the
reader with every opportunity he himself had to evaluate it; while if the evi-
dence is sound, it comes honourably by way of the integral human witness,
demanded also in the Shar'i law courts. An incidental advantage, in the case of
Tabari's material, is the preservation of the vivacity of style which seems to
have been cultivated in such reports and which a judicious legal mind such as
Tabari's would probably not have been able to duplicate on its own: despite
its ponderous length and detail, Tabari's story does not become boring, for
almost every report is humanly vivid.

But perhaps a decisive advantage of the verbatim method for Tabari is
that it allows a writer to avoid committing, himself too publicly on any given
issue; indeed, to suggest two contradictory conclusions at once to two dif-
ferent sorts of people. Most readers (even when a work is quite explicit) see in
what they read only what they have expected to see, or at least what does not
depart far from the categories they are used to thinking in; if bare evidence is
presented them, most readers will readily deduce whatever fits most readily
into the patterns they are already familiar with. If they are not prepared to
face certain problems, the writer is not forcing them to do so; but if they are
prepared, the writer gives them the leads they require; thus every reader is
satisfied at his own level. It was important for Tabari to try to please a wide
audience. Teaching his own system of fiqh in Baghdad, he was fully in the
public eye. His viewpoints were sufficiently free that the zealously hadith-
minded faction then dominant in the city distrusted his loyalty to what they
regarded as proper Sunni doctrine, and at one time he barely missed lynching.
Tabari turned the method of hadith reports to masterly account on occasion.

354 THE CIVILIZATION OF THE HIGH CALIPHATE

Tabari on the death of 'Uthmdn: the great divide for the Shari' ah-minded

The events at the death of 'Uthman present a singularly revealing problem
for the pious Jama'i-Sunni historian, even as they present a prima facie vin-
dication for the Shi'is. The Muslim community as an ongoing Ummah ex-
perienced then its first flagrant breach of unity. Not only was unity broken in a
political sense, which might allow the event to be handled as a repetition of the
Riddah wars. Unity was broken within the Medina community, with close
associates of the Prophet lined up against one another and the Prophet's suc-
cessor murdered as the outcome. The foundations of the Muslim Ummah were
called in question. Historically, all later breaches derived from this one. For
the Sunnis, it was a test case: if the whole body of the associates of Muham-
mad was to be relied on as transmitters of hadith reports and, more generally,
as exemplars of the Ummah at its historical best; and if the Ummah was in-
deed to be regarded as specially blessed by God (as implied at the least in the
doctrine of ijma.' consensus), how is such a calamity to be interpreted? It was
indeed the first fitnah, the first trial or temptation, not only for the original
participants, but above all for the later historians, who so named it.

Tabari' s presentation of the death of 'Uthman confronts the dilemma
squarely, in its own way. But this becomes clear only if one understands the
genre form within which he wrote, and examines what he did with it there.
One must attend closely to his sequences. Tabari tells at length of the events
that led many Arabs to be discontented with 'Uthman. Then he announces
that he will recount the first public insults hurled at 'Uthman, and what led
upto his killing. He begins, however, with what seems like aredundant remind-
er of the causes of people's discontent with 'Uthman's weakness: 'Uthman
gives some camels, sent in as zakat, to certain members of his own Umayyad
family; on hearing this, 'Abd-al-Rahman b. 'Awf, a leading associate of
Muhammad, has them brought back and distributed among the ordinary
people— and 'Uthman stays at home doing nothing. But then Tabari does go
on to describe, as he said he would, the first public insults.

From this point, he builds up the crisis. Through a patchwork of anecdotes,
he tells how a delegation came to Medina from the mutinying Egyptians, how
'Uthman used conciliatory words with them, and how they found themselves
being dealt with treacherously (presumably by Marwan, who was 'Uthman's
chief aide); then how they returned and demanded 'Uthman's abdication,
how the leading figures in Medina gave 'Uthman little support (or found their
efforts at urging peaceful acquiescence to the Egyptians' claims frustrated),
how the Egyptians besieged 'Uthman in his house, how 'Uthman sent for
military support from the provinces, how when this became known the
Egyptians broke down the door by violence and burst in murderously upon
'Uthman.

To this point, the tissue of reports from various sources has been essentially
consistent, very vivid, and quite credible as a human document making the

THE SHAR I ISLAMIC VISION 355

actions of all those concerned intelligible. But before actually going into de-
tails upon the act of killing, at this dramatic point Tabari interrupts his
patchwork narrative with a new account which starts over almost from the
beginning, all derived from one reporter, Sayf b. 'Umar. Sayf's account pre-
sents the mutineers as rabble, inspired by a converted Jew with weirdly here-
tical ideas, animated by base personal feelings; whereas it presents all the
leading associates of Muhammad at Medina as actively supporting 'Uthman
('All, for instance, orders his sons to stand guard at 'Uthman's house) — even
Muhammad b. Abu-Bakr, son of the first caliph and rabid opponent of 'Uth-
man, is made to repent of his opposition (in direct contradiction of a report
given earlier, in which on the same occasion he is merely relatively moderate).
This account explains the absence of effective resistance by any save 'Uth-
man's special family and partisans by saying that to avoid fighting among
Muslims, 'Uthman piously sent his other defenders away; and it presents the
actual killing as full of gross impiety. In contrast to the vividness of what has
preceded, Sayf's account is not only incredible in the various motivations it
implies, but is even relatively schematic in some of the detail of its narrative.

When this account is completed, Tabari resumes his patchwork of less well
concerted reports from the point where it had left off, telling of the actual
killing. He concludes by pairing, in retrospect (and hence out of time se-
quence) , two well-turned speeches : one is by 'Uthman defending himself against
accusations of major crime, reasserting his right to rule, and warning that if
he is killed the community will never again be graced with a genuine united
allegiance; the other speech is by the mutineers, arguing lucidly that whether
'Uthman is personally guilty or merely too weak to control his aides, for
justice' sake he must either resign or be deposed, i.e., killed. Between them
the two speeches present a clear dilemma — the dilemma of how political power
can at once be held within the limits of justice and yet retain sufficient supre-
macy and independence to be genuine power.

This whole section can be read in at least two ways. A loyal, but somewhat
naive, Jama'i-Sunni will accept the central and most obviously coherent
account, that of Sayf, for it solves the historical dilemma for him very nicely:
all of Muhammad's associates were really at one with 'Uthman; it was a com-
bination of 'Uthman's excessive piety with the confusion caused by alien
troublemakers that did all the damage — and the breach in the community was
all a mist ake, in which none of the guarantors of the hadith of the jama'ah was
really at fault. Such a Sunni can dismiss the other accounts as alternative
reports which cancel themselves out by their very confusedness. The section
was often so read.

In contrast, a subtler reader will know that Sayf had a bad reputation as a
hadith reporter and will look elsewhere. He may not fully trust al-Waqidt,
whose narrative he will find to be the core of the chief alternative interpreta-
tion ; indeed, in this very passage Tabari puts him on guard against al-Waqidi's
detail: after a romantic incident which al-Waqidi gives, Tabari interrupts to

356 THE CIVILIZATION OF THE HIGH CALIPHATE

give from elsewhere a shorter version of the same incident, which omits the
central detail of al-Waqidi's version, a detail which if genuine no eyewitness
could have omitted; thus providing effective commentary on al-Waqidi's cred-
ibility without a word of comment. Yet the subtler reader will find al-Waqidi's
version thrust at him by Tabari's arrangement. While the woodenness of
Sayf's version — and hence of the conventional Sunni interpretation of 'Uth-
man — is made to stand out by the very position it takes, interrupting the
flow of the passage at its most dramatic moment, al-Waqidi's version is high-
lighted by its arrangement; for to it belong the anecdote at the beginning and
the two speeches at the end, both of which occur out of sequence, and which
together offer food for thought.

The dilemma of how to have power at once practically effective and morally
responsible, presented in the last pair of speeches, is also developed in some
detail in al-Waqidi's anecdotal material, but the speeches bring out especially
one point: that 'Uthman and his opponents disagree on what is the law in his
case, and both appeal to the Qur'an indecisively. Neither speech, on the other
hand, recognizes an alternative to either accepting the injustice that results
from 'Uthman's inadequacy, or deposing 'Uthman and disuniting the com-
munity. Yet just such an alternative in the dilemma was acted out by 'Abd-
al-Rahman in that first anecdote — the anecdote which attracts attention by
its failure to recount what Tabari says he is about to recount; for, by taking
personal initiative, 'Abd-al-Rahman caused justice to be done in the case of
the camels without questioning 'Uthman's position as caliph.

The initial anecdote and the final passage are complementary. In the initial
anecdote, we find foreshadowed the sophisticated Jama'i- Sunni response to
the dilemma: the principle that every Muslim has the duty of 'commanding
the good and forbidding the evil', so far as he can do so effectively, if no one
else is fulfilling the duty — so that, in principle, the community should not be
entirely dependent on the caliph for justice. Then in the paired speeches we
are reminded of what has to be done before such a principle can be socially
effective in practice — for obviously Tabari is not inviting every Muslim to
behave with the freedom of 'Abd-al-Rahman. The law must be worked out so
that everyone will know what it is: what will make up for the inadequacies of
the caliphs will be the independently developed Shari'ah, upheld by a res-
ponsible Muslim population.

Tabari does not even pretend to give here an exact reconstruction of events.
All his main sources had weaknesses of which he was himself aware, and per-
haps he could not hope to sort out the actuality in detail. Yet at Tabari's hands
the death of 'Uthman, supreme occasion for fitnah, trial, in the Muslim com-
munity, becomes a perfect occasion for showing the naivete of a conventional
response and suggesting, instead, what the true Shar'i response must be. 14

'* Many scholars have underestimated Tabari's own work, looking on him only as a
remarkably useful source of earlier material. His history has been accused of being ill
arranged as compared with his commentary on the Qur'an, and he has been blamed for

the shar'i islamic vision 357

The historical work of Tabari is not, indeed, typical; it is the work of a
master. What it demonstrates is the sort of intellectual subtlety that the
Shari'ah-minded methods made possible at their best. Other historians —
whether as annalists or as collectors of biographies — used the same technical
forms, but their isnads were more simply documentary, their selection re-
flected more frankly their generally more partisan and less sophisticated view-
points. Even among them, however, the technique of isnad documentation as
well as the overall sense of proportion in judging what was important to the
Ummah of Muhammad assured a notable sobriety, factuality, and dignity to
the historical work of this school. 15

The same attitudes prevailed in the study of grammar and literary criti-
cism. These formed, in effect, a single activity, the purpose of which was to
maintain high and pure standards in the use of the Arabic language, which
was the language of Quran and hadith. The discipline originated among
circles whose concern was the precise understanding of old Arabic words found
in hadith and not readily understood by later generations. As in the develop-
ment of the Shari'ah law proper, it was hoped to avoid bid'ah, innovation, and
to preserve a pristine simplicity and homogeneity of attitude. However, the
peculiar importance assumed by this branch of studies resulted from other
currents in the society — from ways of life associated with the 'Abbasi court
rather than with the pious 'ulama.'.

More tenuously, the Shari'ah-minded attitudes penetrated even to the
realm of abstract thought — of theology and philosophic analysis generally.

popularizing Sayf's legends, notably in his account of the conquests — where Sayf tho-
roughly confused the chronology. As prime source for subsequent chroniclers, Tabari
played a role somewhat comparable, in setting attitudes to early events, to the role of
al-Shafi'l in law. Like al-Shafi'i's, Tabari's work achieved ideological clarity at the expense
of historical openness. Yet Tabari's voluminous narrative, though often not comprehen-
sive, is always illuminating, never dry. Surely it was not for nothing that he pointed out
to his students that he was selecting only one-tenth of the reports he could have included
{the common notion that he actually wrote a longer version of which this history is but an
abridgement is not supported by the original wording). A word to the wise should have
been sufficient to put them on their guard. It is no accident that it is only since the dis-
covery of Tabari's work (as Gibb pointed out) that modern historians have begun to be
able to reconstruct the periods he dealt with. Tabari meant it that way.

15 The study of Islamicate historical writing has only begun. Bertold Spuler has
shown, in 'Islamische und abendlandische Geschichtsschreibung', Saeculum, 6 (1955),
125-37, that pre-Modern Muslims did not have so potentially universal an outlook as
Modern Westerners, and has listed much bibliography; but his article is marred by
several false stereotypes and oversights. Franz Rosenthal has written A History of
Muslim Historiography (Leiden, 1952), chiefly focusing on a few Arab chroniclers,
especially of Mamluk times; it is philosophically obtuse, and not very illuminating; see
also his study 'Die arabische Autobiographie', Analecta Orientalia, 14 (Rome, 1937),
1-40. D. S. Margoliouth, Lectures on Arabic Historians (Calcutta, 1902), is a more
interesting but still superficial review. Most useful, though still slight, is H. A. R. Gibb
on 'Ta'rikh' in the Encyclopaedia of Islam supplementary volume, reprinted in Studies
on the Civilization of Islam (Boston, 1962). History of dogma has been studied by
Helmuth Ritter and others, but chiefly only from the viewpoint of the interrelation of
sources, a task Claude Cahen, in La Syrie du Nord (Paris, 1940), has undertaken for
certain chroniclers of the Middle Periods,

35^ THE CIVILIZATION OF THE HIGH CALIPHATE

(We will devote special attention to this later.) Here the concern of the Shar'i
'ulama' scholars was to defend — with appropriate means — the doctrinal posi-
tions the holding of which had been established as legally correct in the fiqh.
Here several mutually hostile factions were engaged, from their differing
points of view. It could, indeed, be legally shown, by the example of Muham-
mad, that God approved the holding of certain opinions, and for many of the
pious this in itself was sufficient. Nevertheless, a great many of the 'ulama'
felt it incumbent on them to show that the positions sanctioned by the Qur'an
and by hadith reports were reasonable; the effort to do this led them into the
whole world of abstract thought.

Finally, the Shari'ah-minded attitudes had a reflexive effect in the very
field from which they had received a major component of their original im-
pulse: in personal devotion and piety. Muslims' sense of cosmic self-orienta-
tion and their ways of responding to divine challenge were deeply coloured by
the Shar'i spirit, however much they varied otherwise.

Muslim Personal Piety:

Confrontations with History

and with Selfhood,

In the courtly society, the surface of life was brilliant and decorative. We have
seen how it fostered the ideal of adab, of polished personal refinement. A culti-
vated discipline of the externals of living — grandly proportioning the build-
ings one lived in, using delicate colour and form in one's utensils and in one's
clothing, above all skilfully refining the words one used and the ideas one
presented — all this could establish a pattern in the daily round which at best
could glow with true beauty. The magnificent court sifted out the loveliest
from the offerings of a world of competitors, and in the protection of that
court's might, and in the prosperity it fostered, such a culture proliferated.
Throughout the empire, whoever could afford such things was inclined to
follow the lead of the court. But set over against this tasteful surface was the
deeper and more tumultuous realm of spiritual responsiveness, expressed in
personal piety. For within almost every man or woman, even among the
privileged, was a rebellious spirit inclined to smash all this elegance in the
name of ultimate reality.

The glitter of the court and its refinement were founded at last upon pride
and greed, upon torture and murder, upon innumerable falsehoods of word
and deed. Nor could any privileged circles in Islamdom fully escape a like in-
dictment. There were those who longed to break through the everyday round
of life, however beautiful, to confront the realities of the universe in the deepest
realities of their own beings, to confront its awesomeness with their own im-
mensity of hope, and find a radical commitment which should claim the stakes
of life and death. Some individuals devoted their whole lives to such an effort;
many others, happy to cultivate the surface as best they could, nonetheless
supported the more committed ones sufficiently to make them a force in the
world. Thus personal spiritual concern became one of the most active forces in
the high culture of the Muslim cities.

The splendour and power of the Muslim state could seem to be the splendour
and power of Islam, and all that a loyal Muslim could reasonably hope to see

359

360 THE CIVILIZATION OF THE HIGH CALIPHATE

in this world; but it was perhaps especially Muslims, aroused by the fresh
Islamic challenge, that felt the thrust of cosmic discontent. Yet they were not
willing to go back to the old parochial systems of piety, nor even to Mani-
cheanism; for it was Justin Islam that theyfound thegreatest impulse to their
seeking. Amidst the wealth and in the varieties of opportunity provided by
the great monarchy, these seekers of cosmic commitment also, like the seekers
of learning or of beauty, found wide scope for inventive initiative. The great
venture of Muhammad and of the early Muslims still had implications that
had not been worked out, a constant stimulus to those for whom the Qur'an
had become a sacred text. Its latent challenges roused many to try to win
through to their aims.

We are speaking here not of religion generally but of personal piety— that
is, a person's spiritual devotion: his manner of response to the divine, to what
he finds to transcend the order of nature, to a felt cosmic dimension of life
giving it ultimate meaningfulness. 'Religion', as we have noted, includes all
the diverse ramifications of those traditions that are focused on such re-
sponses. Religious communities commonly possess not only ultimate responses
but organizations, members of which may have very little spiritual piety even
when they are very loyal to the organization; or they may possess art forms,
or roles in the social structure, or cosmological doctrines. These may reflect the
orientation of the personal piety, but they also reflect other social and intellec-
tual traditions. Nor can piety be reduced to ethics, though it may issue in
special standards of behaviour toward one's fellows. Piety cannot even be
identified with zealous acceptance of myth and ritual, which may occur with-
out real spiritual feeling, and at best may be only partial or occasional expres-
sions of it.

Personal piety is in some ways but a small part of religion. Yet it is the core
of it. For it is in personal devotion (whether by way of the usual rituals, or
otherwise) that the cosmic dimension is entered upon which makes religion
religious; and hence that the whole structure of a religious community ulti-
mately justifies itself. Accordingly, what we call personal piety or devotion
plays a key role in civilization as a whole, at least wherever religious tradi-
tions are of major importance. Changes in culture generally and changes in
moods and styles of piety are closely interdependent. Styles of piety are even
more elusive to study than styles of art, but may be a good deal more impor-
tant.

In some individuals, personal piety is a quiet background to ordinary cares
and pleasures; in others it can be an existential experience and a driving force.
In either case, it has consequences of its own, to be distinguished from the
social consequences of allegiance to a particular religious community. Similar
devotional attitudes may arise within two different religious traditions, or the
like may occur independently of any particular religious tradition. By the
time of al-Rashid and al-Ma'mfin there had arisen within the Islamic tradi-
tion several sharply contrasting styles of personal piety and devotion. No one

MUSLIM PERSONAL PIETY 361

of these styles of piety can be identified simply with 'Islam', though the ad-
herents of each claimed it presented the only true Islam.

The various forms of devotional response of the time, especially within the
Muslim community, jostled and enriched each other. As they interacted with
various social interests and pressures of other sorts, they gave decisive twists
to the fate of the great monarchy and of the proud social and intellectual life
which it fostered. But the various forms of piety that arose at that time re-
quire especially close attention because they entered pervasively into the
whole fabric of Muslim life from that time on; elements of them often persisted
long after the particular tradition that developed them had been eclipsed or
dissipated.

The relation of style of piety to religious allegiance

Devotional response is inevitably a highly personal thing. As in the case of
aesthetic appreciation, every individual has his own bent. The piety of some
types of personality runs to luxuriant expression of a sense of divine grandeur;
in others, a daily walking in the divine Light issues chiefly in a gentleness of
touch in personal relations. Whatever the religious allegiance, all the various
types of devotional expression tend to reappear in those personalities to whom
they are congenial. Devotional response, if it be genuine, varies infinitely as
persons do.

Nevertheless, the forms of devotional response do vary also from religious
tradition to religious tradition and from culture to culture and even from
class to class. It is not simply ritual and creed that vary, but the style of per-
sonal devotion itself. A parallel is to be found in the case of art. Despite all
personal differences, the art of the Italian Renaissance, for instance, shows a
distinct style. No doubt the personal bent of many artists was hampered by the
standards of expectation then fashionable, and their potential genius was left
undeveloped. At the same time, a considerable range of temperament was ex-
pressed even within the broad lines given by the style. So also in piety and
devotion. Presumably, some spirits are relatively cramped in one tradition or
one period, and relatively encouraged to express themselves fully in another,
even though within each style of piety the range of personal variation is con-
siderable. Islam, for instance, like Protestantism, has discouraged the celibate
monastic life. In both traditions, individuals who as Catholics or Buddhists
would surely have become monks have found other forms for expressing their
devotion; but these forms are likely to be relatively unsatisfying to them and
in any case strongly coloured by the prevailing anti-monastic piety.

Not only overall religious traditions, but the various subtraditions within
them have developed distinctive styles of piety, within which personal dif-
ferences must find expression. These styles remain within the broad lines set
by the tradition as a whole, yet may vary drastically among themselves even
so. Thus Christian piety has always been founded upon and coloured by the

362 THE CIVILIZATION OF THE HIGH CALIPHATE

experience of personal redemption. But this has taken very different forms in
the Catholic rite of the mass on the one hand and in evangelical Protestantism
on the other; and within Protestantism, the rigorous and intellectual preach-
ing of sixteenth-century Puritans was very different from the later 'enthusi-
asm' of the American frontier revival movements. The same sorts of diversity
within certain broad lines have occurred in Islam.

Not only different social levels and different historical circumstances may
occasion such different styles; the play of dialogue in a tradition will ensure
that different temperaments receive expression even within a single milieu.
Generally a given cultural tradition, and more especially a religious tradition,
tends to be first established among people of a more or less common tempera-
mental cast. But if the tradition retains the loyalty of a broad sector of a
population, then in the second or third generation, as all temperaments reap-
pear among its heirs, the tradition is likely to be reinterpreted in all the direc-
tions that the full range of human temperament might suggest. The same
formula, which holds the loyalty of all, will come to carry quite contrary
meanings in different quarters. The integrality of the tradition puts limits to
such a process, so that not all temperaments ever become equally at home in
any given tradition, but the process normally allows for sufficient differences
to develop, to strain the unity of any group that attempts a rigorous internal
discipline. The differences in styles of piety that developed among Muslims in
the High Caliphal society reflected in part different social levels and relation-
ships to power. But even more they reflected different temperaments, and the
major groups that formed can be described as reflecting different ways of
seeing life which recur in almost any society.

Piety among Muslims in High Caliphal times was highly diverse; neverthe-
less, certain movements stand out as formative in the course of events. Such a
movement as the Manichean crossed confessional lines; but the most active
movements flowered within the framework of Islam. Among Muslims it was a
time of ferment, in which new patterns were being created — a fact not sur-
prising, since Islam itself, as a religious allegiance, was adding numbers so
rapidly in this time, drawing upon peoples of many different backgrounds.
Like many times of ferment, it perhaps favoured fresh creativity more than it
did profundity. But we may distinguish two main types of piety within Islam,
despite the diversity. One type was the mystical, not yet dominant at that
time. The other was kerygmatic, focused on history. Even more than usually
in the Irano-Semitic traditions, Islamic piety reflected a strong historical con-
sciousness; a kind of consciousness that was becoming rare then in the non-
Muslim traditions, which had been abandoned in favour of Islam by the most
historically active classes.

We have defined the term 'religious' (in the Prologue) as applying, in the
first instance (i.e., as core of the heterogeneous phenomena by extension called
'religion'), to any life-orimtational experience or behaviour in the degree to
which it is focused on the role of a. person in an environment felt as cosmos; a

MUSLIM PERSONAL PIETY 363

focus which has normally entailed some experience of the numinous and/or
some notion of cosmic transcendence, and efforts to respond thereto. One must
add that the devotional or 'religious' response, in this central sense, can occur
in at least three modes — each of which has played a role in Islamic piety. For
we may distinguish three components in devotional religious experience and
behaviour; these components are not mutually exclusive — indeed, they pre-
suppose each other — but they mark different moments of spiritual experience.
Each of these components may be determinative in a devotional tradition, or
even in an individual devotional life, and the other two subordinate to it; and
to the extent that it is so, that component determines the overall mode of the
devotional experience and behaviour.

We may refer to the paradigm-tracing component in personal piety, when
ultimacy is sought in enduring cosmic patterns, in recurrent nature (including
social nature): through myth and ritual as symbolic or interresonant para-
digms, the persisting natural (and cultural) environment may be articulated
as cosmos. For instance, as the worshipper faces Mecca in the mosque and
bows, he sets himself symbolically in the right relation to God — submission;
and to the other Muslims — all facing the same way; and so ever again restores
some cosmic harmony to his life. (Some writers speak as if this were religion
par excellence.) '

Secondly, we may refer to the kerygmatic component, when ultimacy is
sought in irrevocable datable events, in history with its positive moral com-
mitments. In response to a revelatory moment, the environment, particularly
historical society as it is and is about to be, may be seen as radically other
than what it appears, and the individual is challenged to find fresh ways to
respond to its reality. For instance, as the worshipper recites the Qur'an he
may realize that the great of this world are about to die and be judged and are
not deserving of all the reverence they receive; and that he himself must find a
way to change his cringing ways to them and be bounden to God alone. This

1 This is the mode primarily discussed by Mircea Eliade in Le Sacre et le Profane
(Paris, 1965), translated into English from the German as The Sacred and the Profane
(New York, 1959). The illuminating study of religion by Clifford Geertz, 'Religion as a
Cultural System', Anthropological Approaches to the Study of Religion, ed. Michael
Banton (New York, 1966) likewise defines actually only the 'paradigm-tracing' mode —
interpreting the other components in terms of the 'paradigm-tracing'. Incidentally, it
also includes implicitly a wider range of phenomena than what we have called 'religion';
for what he says applies to the whole of what I have called the 'life-orientational' — for
instance, it would apply with little or no modification to the 'paradigm-tracing' dimen-
sion of Soviet Communism (including 'socialist realism', etc.). If one is looking for a
purely formally identifiable cultural 'perspective', as he is, this is proper; and what we
have called the 'religious' represents only a matter of degree within what he has shown
to be the formally more exact category of the life-orientational. This is the case, it may
be added, whether the religious is in the paradigm-tracing, the kerygmatic, or the
mystical mode: something corresponding to each mode can occur in non-religious life-
orientational behaviour, that which puts little emphasis on the role of the person in a
cosmos. (We may perhaps refer to 'self-penetrational' behaviour — which can include,
for instance, certain types of psychotherapy — as the broader category within which
mystical behaviour is the religious form.)

364 THE CIVILIZATION OF THE HIGH CALIPHATE

kerygmatic component has been crucial to the prophetic monotheistic tradi-
tions (and hence some theologians have even set them off as transcending
mere 'religion' taken in its paradigm-tracing mode).

Finally, we may refer to the mystical component in personal piety, when
objective ultimacy is sought in subjective inward awarenesses, in maturing
selfhood: exploring or controlling his consciousness, the person may penetrate
into or through his self to find ever more comprehensive meanings in the en-
vironment. For instance, as the worshipper meditates on God's creative power
and his own pettiness, he may come to feel his body as but a speck floating
among specks; his aims that had mattered so seem silly, and just doing what
God may require seems obvious and easy.

Among most Muslims, to whom religion was of secondary interest, devo-
tional behaviour was commonly in the paradigm-tracing mode; that is, the
latter two components tended to be subordinated to the paradigm-tracing
component: to the extent that the person recognized a challenge in the Qur'an,
or underwent a transformed awareness of himself, this served chiefly to rein-
force the paradigm-tracing experience as such, especially as it came to be em-
bodied in the Shari'ah as universal model. But the great traditions of Islamic
piety were created and developed mostly in either a kerygmatic or a mystical
mode — -they expressed response most characteristically either to a keryg-
matic sense of history or to a mystical sense of selfhood. The formative figures
in Muslim spiritual life, that is, most often worked in one of those two modes;
and to the extent that ordinary individuals deepened their spiritual conscious-
ness, they also tended to one or the other of the two modes.

A. KERYGMATIC ORIENTATIONS

In the High Caliphal Period, the most prominent traditions of personal piety
were marked by a kerygmatic orientation: their positive commitment to
moral challenges revealed in datable events was decisive for the structure of
their religious consciousness and behaviour. This does not mean that para-
digm-tracing components and even some mystical elements were absent from
these traditions; indeed, the paradigm-tracing component was sometimes of
first importance. But in all these traditions, the kerygmatic element played a
role disproportionately large, at least as compared with most religious life of
the Agrarian Age.

The piety of the Piety-minded factions: its exclusivity

Already among the Piety-minded opposition to the Marwanids, many points of
view had been expressed in matters religious. From early 'Abbasi times the
names of multifarious schools of thought have come down to us, or at least of
partisans of one or another leading tenet; it is hard sometimes to sort out

MUSLIM PERSONAL PIETY 365

which names reflect real differences in basic devotional mood. 2 Yet they had
in common an intense sense of the historical challenge of Islam as this was
embodied in the Qur'an. Often even the differences in devotional style among
Muslims were directly related to the common background of factional tradi-
tions which grew out of the Piety-minded factions of Marwani times; the dif-
ferences reflected their common stress on the social and historical dimension
of human life. They all saw the divine power as being expressed in the fate of
human communities generation by generation, but saw it differently. Thus a
political disagreement about the imamate could be explored and yield pro-
found consequences on the most personal level.

This social and historical orientation continued to colour the piety of all
the movements derived from the Piety-minded factions. Often it gave it an
austere temper, demanding a rigorous standard of public decency free of
luxurious display or of other concessions to aristocratic culture that might be
regarded, from an egalitarian viewpoint, as degenerate social corruption. In
general, it made Islam an excellent vehicle for the social concerns of the con-
tinuing monotheistic tradition.

While some older traditions were being attenuated — or enriched — into pri-
marily peasant cults with a tendency to centre on nature myth, the old search
for the combination of a pure personal life with a just social polity was being
taken up above all among the Muslims. Among them, accordingly, the Irano-
Semitic populism, with its personal moralism, came to a singularly consistent
flowering. 3 One side of this populism was expressed in an extraordinarily in-
tensive sense of the moral exclusivity of the true community, which had accep-
ted allegiance to the true creative moment — to the true moment of revelation
and to what it revealed. Only in that community was there truth and validity;
but whoever shared in its allegiance was by that fact not only socially but cos-
mically on a plane above those who refused allegiance, on a plane where the
only true difference among the faithful was in degree of piety. The Muslims
felt themselves the defenders of the faith of Abraham in the midst of re-
paganized dhimmi communities.

This exclusivity was mirrored in the shahadah declaration, made up of two
equally important statements: there is no deity but God, and Muhammad is
the messenger of God. In principle these phrases could be — and on rare oc-
casions were — interpreted as on different levels: the first phrase as universal,
the second as a special case; then any other prophet's name could be substi-
tuted equally well in the second phrase without invalidating the phrase as a
testimony to the essentials of divine truth. The status of the other prophets

2 All too often every possible shade of grouping, from an isolated opinion to a sub-
stantial school of thought or even an organized religious body, is called by the scholars a
'sect' without any distinction of kind or degree or grouping. This can lead to absurd con-
fusions. (Compare the Appendix on Usage.)

3 I have attempted to develop this consideration in detail in 'Islam and Image', His-
tory of Religions, 3 (1964), 220-60; the sense of development over time is largely wanting
there, however.

366 THE CIVILIZATION OF THE HIGH CALIPHATE

could not be denied. But in practice, pious Muslims could not acknowledge
that the traditions derived from the moments of revelation granted to those
other prophets had more than a limited legal validity as compared with the
tradition arising from the revelation to Muhammad. The others were all quite
hopelessly corrupted. Hence the second phrase carried a force not fully indi-
cated in its bare words: it was not just messenger ship, of which Muhammad
was the present instance, but Muhammad' s messengership that was the second
eternal verity to put on a line with God's unity. The messengership of former
prophets was but a pale corollary of Muhammad's, inasmuch as he acknow-
ledged them. There was a sternness and uncompromising dignity in the position
of the Muslim as such, which every Muslim could share in: he alone, in contrast
to all the infidels, had undertaken the trust, had undertaken to be God's re-
presentative in the earth; the others were not really even fully human. If the
Muslim was not in this period the actual political ruler, yet he was equal
member in a community that did rule, and cosmically he shared its status.

The same exclusivity was expressed in the overwhelmingly central role
played by the Qur'an in Muslim piety — by that Qur'an which was in some
measure, at least, on everyone's lips, and which no educated or specially
trained elite could really monopolize. The whole of pre-Islamic monotheistic
experience and lore was drawn on to build Islam, which thus could be defined
as Irano-Semitic monotheism in its most populistic form; yet nothing was
allowed to derogate from the place of the Qur'an as the point of commitment
in the tradition. Qur'anic passages could, indeed, be interpreted away, and
non-Qur'anic viewpoints could be read into the Qur'an. For instance, for one
who read the Qur'an without preconceptions, there could be found in it little
objection to the painter or sculptor but much to the poet; and yet by High
'Abbasi times Muslim piety had generally reversed this judgment — to the
point that the very words which in the Qur'an had designated merely special
symbolic stones were being transferred to the sculptor's art, so that later
generations saw their latter-day prejudices retrospectively embodied in the
Qur'an itself. Yet for all that, the central challenges of the Qur'an were
strongly felt and the Irano-Semitic heritage was closely canalized through that
document. Islam as felt by the more pious, then, could also be defined as
response and loyalty to the Qur'an and its message.

The potency of the Qur'an

The Qur'an's appeal proved to wear well. What had challenged the old Arabs
of Mecca and Medina carried perhaps as potent a challenge to the city mer-
chants and craftsmen of the settled lands from Nile to Oxus. As has happened
in many religious traditions, the relative lack of intellectual sophistication of
Muhammad's environment, which was evident throughout the Qur'an, lent
itself to highlighting the elemental realities of human living. The homely situ-
ations, the everyday analogies could be understood directly even by the

MUSLIM PERSONAL PIETY 367

illiterate. Perhaps more important, there was no preoccupation with the subtler
niceties of either aesthetic or moral awareness as cultivated in a learned elite,
to disguise either the savagery or the nobility that men have in common. For
chopping off the hand of a thief, the Qur'an offered neither subtle apologies
nor (as were its commentators to do) niceties of legal circumstance; the cruel
punishment stood forth as a judgment, within the terms of awareness of ordin-
ary people, both on the thief and on whoever would allow either his pity
to blind him to responsibility, or his wrath to yield to an arbitrary vengeance.
Even those who feel they have found a better way than was available to the
Medinese to respond to the demand for justice so presented must pause to
consider whether their response is truly as soberly balanced as that, or whether
perhaps it is tainted with sentimentality or even self-righteousness.

For the Qur'an continued, as in Mecca and Medina, to be a monumental
challenge. In its form, it continued, even after the ending of active revelation
with Muhammad's life, to be an event, an act, rather than merely a state-
ment of facts or of norms. It was never designed to be read for information or
even for inspiration, but to be recited as an act of commitment in worship;
nor did it become a mere sacred source of authority as the founding of Islam
receded into time. It continued its active role among all who accepted Islam
and took it seriously. What one did with the Qur'an was not to peruse it but
to worship by means of it; not to passively receive it but, in reciting it, to
reaffirm it for oneself: the event of revelation was renewed every time one of
the faithful, in the act of worship, relived the Qur'anic affirmations.

Accordingly, the worshipper reaffirmed for himself through the Qur'an, in
whatever passage of the Qur'an he was uttering, its single massive challenge:
the challenge best summed up in the word tawhid — the assertion of God's
unity. He certified anew that the authority of the Creator-god and His de-
mands on human consciences confront us without any lesser rival, any inter-
mediate source of norms, any slighter duty; thus he undertook to live up to a
standing claim which every individual faced anew in the Qur'an each time he
renewed his recitation of it; a demand to which he rededicated himself in every
act of worship. Every verse of the Qur'an presented and illuminated in its own
fashion this challenge, applied to numerous details of common life or envis-
aged through the lessons of nature and of history. 4

1 There is pressing need for a study of the Qur'an from a modern scholarly viewpoint
that can provide a way of reading the Qur'an alternative to that traditional in the West.
Even if we get a better translation than yet exists in English, we will need an extensive
'comp anion to the Qur'an' that will show the reader what to look for. At present, what is
available to the general public is typified by W. T. de Bary, ed., Approaches to the
Oriental Classics (Columbia University Press, 1959), devoted to ways of introducing
exotic classics into Western general education. There, Arthur Jeffery, a philological
authority on the Qur'anic text, says that what commends the Qur'an to general study is
solely its enormous prestige among Muslims; he concedes it neither religious nor literary
value. Its prestige he tries to explain as resulting from its unique theological status, but
he does not say that that status itself must be explained — that the Qur'an must have
won considerable prestige on internal grounds before serious persons would be willing to
assign it so unique a theological status.

368 THE CIVILIZATION OF THE HIGH CALIPHATE

The unique potency of the Qur'an, calling for a person's undivided atten-
tion to its single and total challenge, brought a purely religious component to
the growing movement of iconophobia in the Irano-Semitic traditions; a
movement which typified their piety and how that piety was fulfilled in Islam.
The use of images could be felt to be inappropriate to a prophetic religion of
the moral God because of its association with the nature gods; but in the
presence of the Qur'an they became directly distracting and divisive, quite
apart from any such associations. For the Qur'an itself can serve as a sensible,
almost tangible symbol of the One whose challenge it presents.

That challenge is single-mindedly a moral one. If, in the Qur'an, we are
directed to the glories of nature, it is not that we may praise God's beauty or
stand in awe of His wisdom, but that we may be warned of His power to en-
force His ordinances. In the spiritually more sensitive individual, the ex-
clusive focusing of his thoughts on the Qur'an could generate an overwhelm-
ing moral force that might mould his whole personality. Hence to juxtapose
any other symbols in worship alongside the Qur'an, however honestly they
might point to other aspects of divinity, must necessarily, in the nature of the
power of symbols in human beings, share in, channel away, and finally dissip-
ate the concentrated devotional energies. Such alternative releases of the
emotions were not alternative means of coming before the One; rather, they
divided and weakened the devotion to the One expressed in the Qur'an, and to
its moral demands. It may be said that the doctrine of the unity of God, which
has been so central to developed Islam, is largely the theological expression
of the unity of the act of worship at its best, its undivided dedication to realiz-
ing the moral lordship of God over the worshipper.

Accordingly, the central presence of the Qur'an excluded such symbolic
expression of more limited distinct aspects of the divine-human relation as in
Christianity was given in the sacraments; and with the sacraments, it excluded
the priest-craft which the sacraments presupposed. Necessarily it excluded
those other symbolic expressions of spiritual awareness represented in the
arts; much as later happened in a somewhat similar movement in Protestant
Christianity. But the feeling of the pious went beyond the service of worship
and the place where the worship was held. All life should be informed with the
religious spirit; nowhere should be tolerated anything that could rival the
Qur'an in evoking the deeper responses of the spirit. The whole imaginative
life was suspect: science and fiction, music and painting. So far as any art that
is true to itself is not, in fact, a mere pleasing of the senses but evokes the
whole spirit, all art was potentially a rival to the Qur'an, a subtle form of
idolatry. Nor is science merely an objective satisfaction of curiosity: it calls
for its own morally single-minded devotion. The pious, therefore, could well
fear all the arts and sciences wherever they appeared, and indeed all aspects of
high culture that did not clearly subtend the moral purposes to which the
Qur'an summoned.

Finally, the unity of the cult centred on the Qur'an issued in the exclusivity

MUSLIM PERSONAL PIETY 369

of the religious community itself: if within Islam no rival form of cult could
be tolerated, however monotheistic, then still less could be admitted the legiti-
macy of any religious communities rival to the one which maintained the
Qur'anic cult. The one God implied the one medium of worship and the one
worshipping community.

Shar'ism as an expression of Qur'anic piety

Among the very seriously concerned, this spiritual experience was doubtless
the strongest motive to the exclusivity of Islam; its exclusion of all cult im-
ages, more generally its exclusion of all symbolism rival to the Qur'an and of
all rival communities. Yet these few concerned could not, by themselves, have
enforced even a small measure of exclusivity. Their sentiment could not have
been effective if it had not coincided with the more general tendencies of
populism as fostered in a mercantile milieu : the distrust of the aristocratic, the
luxurious, the exceptional, which must include all true science and art; and
the substitution, for these splendours of high culture in high society, of the
pre-eminence of the sacred community in which the most ordinary man found
his dignity in the daily moral relationships of his life. Hence it was only where
the sentiment of populism at large reinforced the sense of worship of the most
pious, that their spiritual convictions took effect. Accordingly, the most luxu-
rious form of art, requiring the greatest aristocratic or priestly taste and re-
sources, sculpture, was almost entirely banned; while that art which every
class could indulge in, poetry, was almost never condemned. Yet perhaps the
populist impulse itself could not have had so much success, had it not been
supported by the spiritual insights of the exceptionally sensitive; certainly it
was only in the mosque itself, where such men had special prestige, that the
sentiments of exclusivity were fully effective.

In any case, it was precisely what was universal in the vision of Islam, its
hope of equal justice and of a human responsibility under transcendent norms,
that issued in the exclusivity of Islam. The very response to the vision which
allowed that vision to be embodied in a living tradition, and the responsible
commitment which then carried it forward in actual society, were what closed
Islam off from rival values and rival traditions. The Qur'an, to whose words
Islamic symbolism came near to being restricted, became the one great con-
crete image in Islam,

Despite the immediate potency of the Qur'an, the text by itself would still,
presumably, not have been enough to focus the tradition so forcefully without
being tied in concretely with the ongoing community life. In Muhammad's
time it had answered to the current sequence of events, illuminating and
guiding them. Later Muslims had to find the equivalent for their own genera-
tions. One solution, embodying the challenge in daily life, was to work out
rules of practical morality which would be consistent with the Qur'anic cult.
Thus arose the Shari'ah, the autonomous body of law being spelled out for

37» THE CIVILIZATION OF THE HIGH CALIPHATE

increasingly detailed cases. In this process (which never really ceased), the
Qur'an played a central role and so entered actively into the community life.

The Sharf'ah was given further emotional force because it embodied loyalty
to the Muslim community, whose idealized ways it described. The exclusivity
latent in the Qur'an was early complemented by an exclusivity grounded in
the historical Muslim community. In the reaction that followed the third
fitnah and the 'Abbasi triumph, this communal orientation of this Shar'i
spirit was explicitly emphasized: that is, loyalty to the community of Muslim
allegiance, even at the expense of any other value. It was readily combined
with the new Shar'i spirit; since for the Piety-minded, who had rejected the
Marwani leadership and did not love the 'Abbasi, identification with the com-
munity could only mean acceptance of the Shari'ah.

In Shi'i circles, this spirit helped to discipline even the Ghulat theorists.
The imam Ja'far al-Sadiq, soon after the advent of the 'Abbasids, found it
wise to disown a prominent follower, one of the Ghulat, Abu-1-Khattab, for
going too far in neglecting the growing Shar'i spirit; he was more concerned
with inner symbolisms than with legal applications. (Not long after, Abu-1-
Khattab was executed by al-Mansur for heresy.) In general, the thinking of
the Ghulat theorists was curbed about this time, at least verbally, so as not
to offend the Shari'ah-minded. Thus the later theorists among all the Ja'fari
Shi'ahs could exalt their imams as 'proofs' (hujjah) of God or carriers of a
divine light — both of which notions could have broad metaphysical implica-
tions in the Irano-Semitic tradition; but (in contrast to the practice of the
first Ghulat) in deference to the common legal supremacy of the Prophet par
excellence, the imams were never thenceforth called even very minor 'pro-
phets' (nabi) despite the lesser implications of such a term.

A related communalistic mood appeared among the legists themselves.
Abu-Yusuf, disciple of Abu-Hanifah, stressed in Baghdad the legal norms of
the old times under the Marwanis almost as much as did al-Awza'i, leader of
the Syrian school, who was naturally nostalgic for the time of Syrian primacy.
Through such men, something of the older Arabism survived in the Piety-
minded Shari'ah itself.

But even so, this Shar'ism was not sufficient in itself to establish a living
involvement with the Qur'an, and so to form the whole content of a vital mode
of piety. Some sort of Shari'ah-mindedness was almost universal among the
Islamic movements stemming from the Piety-minded factions. But each of
them added its own further ingredients.

We shall discuss here first the sorts of piety that arose among those who
refused accommodation to the established Muslim order — Kharijis and es-
pecially Shi'is: Zaydi Shi'is in their persistent politically-minded insurgency,
and the chief radical Shi'i groups, Twelvers in their chiliastic longing and
Isma'ilis in their esoteric and conspiratorial intellectualism. Then we shall
return to those who accepted the j ama'ah as established, notably the moralistic
Mu'tazilis and their rivals the populistic and loyalistic Hadith folk, chief heirs

MUSLIM PERSONAL PIETY 371

of the older Piety-minded groupings; and especially the Sufis, whose mysti-
cism formed a primary basis for most Muslim devotional life in later centuries.
I shall necessarily present here what I do see fairly clearly in these several
traditions, and omit what I do not see; hence what I say will look better in-
formed than it is. But it is only with closer studies of the piety both of indivi-
dual leaders and of more typical individuals in each tradition that we can
hope for an adequate understanding of what is now, all too often, a field for
conjecture.

We have seen how by early 'Abbasi times a distinction was forming between
the Shi' ah, the partisans of 'Alt, looking to a repurification of Islam, and the
Jama'ah, the partisans of community solidarity, looking to maintain what had
been achieved; the two positions appealed respectively to persons more inclined
to risk all for an ideal, and to those more inclined to estimate that no better
could be had than was had. Within each of these broad allegiances appeared
further distinctions — reflecting again (within each group) a relatively idealistic
demand for perfection, or a relatively practicalistic demand for conservation;
or reflecting, further, a distinction between those more inclined to public,
exoteric truth accessible to all, and those inclined to an esoteric truth acces-
sible only to an elite. Out of diverse combinations among these and yet other
tendencies arose several strong movements with a distinctive style of piety,
some of which played major historical roles.

Two early movements, which persisted long into 'Abbasi times, especially
represented the spirit of austerity present among the Piety-minded dissidents.
They demanded a turn to higher standards than generally prevailed, and de-
manded this of the M uslim public at large, with little condescension to human
weakness. They gradually became less prominent during High Caliphal times,
and later persisted only in isolated places; but as long as a strong central em-
pire still presented a relatively simple political challenge to the rigorous re-
former, they offered cogent answers for certain temperaments.

The movement most notorious for its social austerity, that of the Kharijis,
had rejected even the Shi'ah as a vehicle of reform. We have noted them as
working out a consistent attempt to maintain the absolute equality and re-
sponsibility of believers within a homogeneous community, often at the ex-
pense of abandoning the Muslim community at large and retiring into self-
righteous war bands. They were intensely concerned with Shar'i questions, but
insisted, above all, on effectively righteous public order. Sometimes they ex-
plicitly tolerated behaviour that other Piety-minded groups banned in theory,
so as to make their theory really applicable in practice. But on the whole their
temper was puritanical and exceedingly militant. The Ibadi Kharijis were
able to establish, among considerable populations in out-of-the-way Muslim-
majority areas (notably 'Uman in eastern Arabia), a public norm at least
superficially proper. After the eighth century, however, they gradually ceased
to be important except in these limited areas.

Within the Shi'i movement, a very like position was represented by the

372 THE CIVILIZATION OF THE HIGH CALIPHATE

Zaydiyyah. The Zaydis insisted, from the time of al-Qasim al-Rassi (d. 860),
their great theorist, that the true imam must be that 'Alid (descended from
Fatimah, that is) who combined mastery of legal and religious teaching with
the political initiative and acumen to carry out armed rebellion against the
authorities. The imam might be of the line either of Hasan or of Husayn, and
need not be the son of an imam; nor need there always be an imam at all, if
none appeared who was truly qualified. The result was a series of highly com-
petent imams — ruling in out-of-the-way places, especially in the Yemen.

Shi' ism, the piety of protest

After the great disappointment to pious expectations at the time of the 'Ab-
basi revolution, Shi'ism developed more and more in its own course apart and
cultivated its own forms of piety. The Shi'is gradually developed their own
sense of the Shari'ah and their own variants of it. But to it they added a
strong 'Alid loyalism of an intensely religious quality. (I use the term 'Alid
loyalism for the varied complex of special religious attitudes associated with
loyalty to the 'Alids — not only reverence for the Alids themselves, but cer-
tain exalted ideas about Muhammad's person and the supposition of a secret
teaching he transmitted specially to 'All, and so on — whether these attitudes
appear among Jama'i-Sunnis or among those who, by explicitly rejecting the
jama'ah, identified themselves as Shi'is in the proper sense.) For the Shi'is, it
was not sufficient for the law to be an autonomous corps of authoritative norms
which the community as a whole would maintain against any given ruler. Its
continuity must be ensured, as in Muhammad's time, through the presence
of an authoritative spokesman for the divine will — a true imam. 'Alid loyalism
offered just that.

The more intense Shi'is had from an early date felt that 'Ali was not merely
one caliph among others, but that he had had a special authority to lead the
community of Muhammad. As they explored the implications of his early
leadership and of their moral commitment to him, they became sure that he
alone had acquired, through his closeness to his cousin and father-in-law, the
Prophet, the full 'ilm necessary to guide the consciences and the lives of true
Muslims — the knowledge of what was lawful and just, and perhaps of many
other things as well. In abandoning him, they felt the Muslims at large had
abandoned the truth; since that time the true adherents of Muhammad had
been the few who remained loyal to 'Ali and his family, in which the sacred
'ilm continued from generation to generation. 5

Such a point of departure necessarily launched an active and highly varied
dialogue among those who felt themselves to be the elite remnant, whose

5 Something of the mood of this 'Alid loyalism emerges in Louis Massignon, 'Salman
Pak et les premices spirituelles de 1'Islam iranien', Soc. etudes iraniennes, no. 7, 1934.
Highly valuable studies by Massignon on ShS'ism and Shi'ism and Sufism are listed in
the 'Bibliographie' in vol. 1 of Melanges Louis Massignon (Damascus, 1956), itself a
rich collection of articles on Islamic religion and other Islamicate matters.

MUSLIM PERSONAL PIETY 373

Islam must be distinguished from that of the common herd of renegades.
Though a certain generalized 'Alid loyalism persisted among the population at
large, who commonly felt that any 'Alid claimant to the caliphate at least
deserved a hearing, those who had assumed a commitment to the 'Alid cause
could not leave it at that. The Zaydis, indeed, developed (as we have noted) a
full-fledged 'Alid loyalism which was sectarian but hardly esoteric.

In contrast to the Zaydi position, which offered just a more purified version
of the ordinary Muslim outlook, the position of the more radical Shi'is, those
committed to a designated imamate of nass as a basis for sectarian continuity,
became widely influential even beyond the circles of those who explicitly
adopted the position. For it presented a distinctive style of piety which ap-
pealed to a widespread human temperament, and which loomed especially
important within the structure presented by any monotheistic tradition. First,
it was an esoteric position: the radical Shi'is looked to a privileged truth in-
herently inaccessible to the unworthy masses, and into which those who were
worthy had to be specially initiated. The notion of a secret, hidden wisdom,
which only an elite were worthy of, was an almost inescapable corollary of the
position that only a few were privileged to perceive the true destiny of Mus-
lims and hold to the true leaders of Islam. Hence the sectarian Shi'is focused
their interests on the special 'ilm of the imams which only Shi'is could appre-
ciate. But the appeal of the esoteric was not enough alone. Other forms of
esotericism were available, in Islam, that proved compatible with a Jama'i-
Sunni position; notably an esoteric approach to the inward personal experi-
ence of mysticism, into which a disciple could be initiated only by an experi-
enced master. What was distinctive in esoteric Shi'ism was that it presented a
privileged vision of history. It was a 'kerygmatic' esotericism.

On the everyday side of common doctrine, urged indiscriminately upon
everyone and hence (from an esoteric viewpoint) exoteric, the notion that his-
tory began and was to end decisively and once for all was fundamental to the
moral position of monotheism. The traditions had all early developed an es-
chatology, a doctrine of what was to happen at the end of time when all
humans would be duly rewarded or punished. (This was implicit in their
kerygmatic mode of piety, as this was expressed in monotheistic terms.) But
this exoteric eschatology need not amount to more than a personal destiny,
in which adherence to the one true historical community will have been im-
portant, but which was not really a part of history, for it took effect only
when history was over. Alongside this milder approach there was always a
more radical strain within the eschatological thinking of the traditions: chil-
iasm. Not merely a final Judgment was expected after history, but history it-
self was to be completed with a blissful millennium, a culminating age when
the world as we have known it will be put right within the terms of history.
Chiliasm expressed radical social protest. Very shortly, the wicked great of
this world would be humbled or destroyed, and the lowly, or those of them
who had proved capable of maintaining the true faith and the true loyalty,

374 THE CIVILIZATION OF THE HIGH CALIPHATE

would be exalted to share the good things of this world free from oppressors.
The world would be 'filled with justice as it is now filled with injustice', in the
Shi'i phrase.

Such an expectation meant not merely a hope for the future, but a re-
evaluation of present social and historical life. In the light of what was to hap-
pen — and every generation found reason to suppose that it was likely to happen
in their own time — present social and political arrangements were temporary
and even highly precarious. It was always possible that the foreordained
leader (the Mahdi) might appear and test the faithful by summoning them,
just as they were, to launch the great social transformation themselves under
his command, with the promise of divine succour when it would be needed. But
even before he appeared, the social role of the various elements in the popula-
tion took on a changed air for those who knew what was to happen. Every
mundane historical event might presage or prepare the Mahdi's coming. The
faithful were always on the alert, ready to take their part in the final acts. In
this way, a chiliastic vision dramatized all history, in the present as well as
the future. But the more seriously a chiliastic vision was received, the more it
necessarily contradicted received social viewpoints. Both socially and intel-
lectually, a chiliastic vision had to take esoteric form, if only in self-defence.

Once its partisans ceased to attempt immediate and direct political action,
the Shi'ah became the chief vehicle among Muslims of chiliastic hopes, per-
petuated in terms of esoteric lore. For Shi'is, the Qur'an had a secret interpre-
tation which made it speak of the imams and their historic fate; and indepen-
dent books of predictions, circulated in the name of an imam, commonly
Ja'far al-Sadiq, darkly foretold events to come in such a way that every new
generation could see its trials and hopes mirrored there. Many Shi'is were also
interested in other esoteric lore — notably the occult sciences, such as chemis-
try in the form of alchemy. But an esoteric historical vision was of the essence
of radical Shi'ism. Even when chiliastic expectations spread into Jama'i-Sunni
milieus, it was largely in 'Alid-loyalist forms that it did so.

This Shi't esoteric thinking built largely on the work of the early Ghulat
theorists. Their speculations had exalted the role of the imam as saviour of the
ordinary human being, and often introduced esoteric notions from pre-Islamic
heresies. Faith tended to be conceived in terms of a personal devotion to a
divinely guided hero, and a heady enthusiasm sometimes resulted. The more
zealous enthusiasts, however, had rarely been long separated from less vision-
ary persons, concerned above all with the true imam as a decider of points of
sacred law and restraining their curiosity for anything more strongly esoteric.
From this varied milieu grew both' the Twelver and the Isma'ili sects, the
Ja'fari Shi'ahs.

The Twelver Shi'ah

Of the radical Shi'i groups, that which was centred, at the time of the 'Abbasi
revolution, on 'All's great-great-grandson Ja'far al-Sadiq, was the only one to

^ * 2

J3
M

CD

B

t^

fa

r

£

■O

"3

bo

s

+j

a

rt

'i>

d

?

ri

'S

J2

f

< +.

£l CO
3^
&•«

2 o

H £•

c^ E-S

'Cfa

•d
d

<T-I

XS
N

XX

H

5/3

D

d.2
43 .□

00 60 ,-,

£■1 «?

O ^ CO

i— J ±! «+-t

(A

3

c*3

d
N

1-

iO
00

ri"

<d

+-» ' — ■
MOO

'C N

S On

$a

3"°

<3

CO >,

- M

_- 2 ri
r5 'S -P

^,3 "5 5.

X>,3

T3 3

d ri
B&

B 8

d o
*S 3

II

ds o

-2-3

-ri
.3

d (

ra S

_- C
■O d

3. en

>>&

p en

I'-

?* CO

J, 3

■S <!2

rt 3

3
d §

S.S
IS

ri< 2— =3-

C H O

w-a •£■

bo
C

c

■as

s s
-as

ri +i

ri __

ori^-

a-§ a s*

b s « ^
S3 "-t-i xs rs

"3 'C r3 ^

? 2 bog
ttSCh
^ «ri '-a U

■^■3 S s

'O ri tfcr 3
4> . h +■
■yo ° bo

<D« ffi.S
« r! ° S

°<"3 3 'bb

^3 M «>
Ph5 ri*

H "C •- m
ri ri'g g

<e& ^

13
ri

a
-s-

ri
J*
3

bo

3

•3

bo g

"3 E

ON*J3

o
■d t3

Sfal

^■5I.B

_ ■ «g rfs -

* rt 3 W

i ri •

13 E >

lofa

3 3 bo

X

<

X

H

n ^3

376 THE CIVILIZATION OF THE HIGH CALIPHATE

have a great future: partly because the movement harking back to Ibn-al-
Hanafiyyah had aborted in the 'Abbasi cause; partly because of the Fatimid
ancestry of Ja'far's line; partly, no doubt, because of its intellectual and
political disciplining at the time of Ja'far. One of the chief sects to issue from
this group traced six more imams beyond Ja'far. Each time an imam died
and a successor was to be recognized, issues were raised about the nature of
the imamate (was the proposed successor qualified?), and consequently of the
group and of its faith, and were resolved through the choice of the next imam,
which was almost never unanimous. Except in the case of the Isma'ilis, how-
ever, the bulk of this group, the Twelver Shi'is, eventually settled on a single
course and dissidents died out.

Ja'far himself had been acknowledged as his father's successor almost with-
out dissent; he was personally learned and respected, and the sectarian view
of the imamate was still barely formed anyway. But at his death questions
arose. At one time he had duly designated his son Isma/il as successor; by the
rule of nass, succession by designation, this should have settled the matter;
but Isma/il died before his father. Was the imam fallible in so crucial a matter
as the designation? If not, had he been dissimulating in foreknowledge of the
event, so as to protect the true heir from the eyes of the authorities — -who
looked on any Shi'i imam with understandable suspicion; or was the designa-
tion merely contingent, as in other human affairs, to be superseded by a new
designation at need? Without necessarily agreeing on the answer, most of the
Shi'is turned to the eldest surviving son, 'Abd-Allah, who seems to have
claimed a second designation by Ja'far. Some objected to him, however, claim-
ing that, on being questioned, his 'ilm knowledge did not show itself sufficient.
Was this a sound test on the part of ordinary human beings? Within a few
weeks, 'Abd-Allah's death without heirs seemed to confirm the doubt about
him. The majority turned to another son, Musa al-Kazim, but without all
agreeing whether he was 'Abd-Allah's successor or had been the true successor
to Ja'far all along; for this question hinged on how one was to interpret
the imam's 'ilm. Many, however, clung to the designation of Isma/il and
proclaimed Isma'il's son; these were the Isma'ilis. Still others seemed to
have maintained that Ja'far himself, dead or alive, was still the only true
imam.

Musa spent most of his time in an 'Abbasi jail and was spared much ques-
tioning by the faithful. On his death, his son 'All al-Rida was accepted, though
not unanimously. But 'All's son at his death was only a young lad, and his son
in turn was still younger, too young to have received instruction from his
father in the family lore, at least in the ordinary way. In what sense were such
orphans to be looked on as still possessing the private 'ilm which Muhammad
had confided to 'AH? Those who accepted them — and not all did — were forced
to acknowledge not merely a divine designation of the given leader as authori-
tative, in the sense that a constitutional high court is authoritative whatever
its decision; they had to acknowledge a more active divine intervention: at the

MUSLIM PERSONAL PIETY 377

very least, to protect the imam from false learning as he grew up; and pre-
ferably in more positive ways: by way of a magic book containing all know-
ledge, which remained in the imam's possession alone, or, more simply, by way
of divine inspiration; or even by way of the imam's metaphysical substance,
that special divine Light which was infused in him and caused him to know
truth by his very nature.

At last Ja'far's great-great-great grandson, Hasan al-'Askari, died with no
known son at all (873). Most found his brother, who had been at odds with
him and now claimed the position, unacceptable. But by this time it proved
feasible and necessary to suppose that Hasan must have had a son and that
that son must have disappeared; presumably being hidden to save him from
'Abbasi persecution. Diverse accounts sprang up to fill in the details; but
what mattered was that a son for whom such total precautions were taken
must be a very special imam. It was he, then, who was to be the Mahdi,
Muhammad al-Muntazar, 'Muhammad the awaited*. Reckoning without 'Abd-
Allah, he was the twelfth imam, and innumerable prophecies were now found
to foretell the greatness of the twelfth of the line. (Hence his adherents were
called 'Twelvers', in Arabic, Tthna'asharis', as recognizing just twelve imams;
but they generally preferred to call themselves 'Imamis'.) He was held to be
waiting in hiding, undying, till at the end of the world he should return and
finally bring victory to his loyal partisans, and truth and justice to prevail in
the world.

The custom had arisen for each imam to communicate with his followers —
for instance, for the purpose of receiving the alms paid in to him as imam, and
of disbursing them for pious purposes — through a wakxl or safir, a personal
representative. (The imam did not always wish to expose himself personally to
his followers' zealous expectations, and in any case was usually either in
Medina, far from Kufah, or else was imprisoned as politically suspect — -if he
was not a child, requiring adult guardianship.) On the disappearance of the
twelfth imam, a succession of four wakils, each naming his successor, con-
tinued this role on a provisional basis, so maintaining the continuity of the
organization and its financial structure. But in 940 the last of these died, re-
fusing to name a successor. Thereafter there was no central organization. The
sect maintained itself nevertheless; the time of the four wakils was called the
'lesser absence (of the imam)', the lesser Ghaybah; thereafter the Twelvers
found themselves in the 'greater absence', the greater Ghaybah. But by then
the High Caliphal state itself was at an end.

The imams, then, came to be invested with cosmic worth, and their lives
reflected the sad vicissitudes of the divine cause among ungrateful mankind.
As 'Ali had been abandoned by the Muslims and finally murdered, so the
Twelvers came to feel that each of the imams that had followed him had been
persecuted and finally executed or at least secretly poisoned by the wicked
and worldly Muslims in power, both Umayyads and 'Abbasids. Muhammad
himself was depicted as suffering silently the incomprehension and infidelity

378 THE CIVILIZATION OF THE HIGH CALIPHATE

of those who seemed to support him most, and anticipating the sufferings of
his descendants and their followers; his daughter Fatimah, mother of 'All's
sons, wept over the wrongs she and her children suffered at the hands of the
violent 'Umar; and 'Ali, too just for this world, received a martyr's death.
All these wrongs the pious Shi'is wept for, but above all for the murder of
'All's son, and Muhammad's grandson, Husayn, betrayed by his own sup-
porters to be tormented with thirst in the desert and slaughtered by his
enemies at Karbala.

Set as a minority against a too-triumphant world, the Shi'is wept on the
anniversaries of their wrongs and at the tombs of their heroes, and believed
that their love for the suffering imams would win them forgiveness for their
own sins and a share in the victory of the righteous in the end. The
resemblance at many points to Christian piety has often been noted, but no
historical link has yet been traced.

As to doctrines of law and social order, the Shi'is differed but little from
others of the Piety-minded. They had their own Shi'i Shari'ah (in points like
inheritance, it clung more literally to the Qur'an than did the Jama'i-Sunni
schools) ; and their own Shari'ah-mindedness in general, from which various
groups tended to deviate less or more. Like the Jama'i-Sunnis, they depended
on hadith reports for validating their tenets ; but they looked to hadith reported
from the imams, and vouched for in the isnad by known Shi'is, in contrast
to the kinds of isnads preferred by Sunnis. Yet except as regards the special
role of 'All and the imams, their hadith were often almost identical with those
of the Sunnis. What gave them their vitality was a special mood of devotion
superadded to the common Shar'ism.

This Twelver Shi'i piety proved highly attractive. With its exaltation of
both 'Alt and Muhammad as superhuman heroes, it has, from 'Abbasi times
on, affected the outlook of Jama'i-Sunnis as well. For a time, in the tenth and
eleventh centuries, the Fertile Crescent and much of Iran were under Shi'i
rulers. Shi'i days of triumph (the recognition by Muhammad of 'Ali as imam
on the day of Ghadir Khumm) and of mourning (especially the month Muhar-
ram, when Husayn was killed) were publicly solemnized. Many of the leading
figures then in the arts and sciences were Shi'is. Yet throughout the period,
Shi'is remained a minority and could not control the overall social life. Sunnis
could not be persuaded to adopt a minoritarian mood of self-pity and self-
blame, such as Shi'ism sometimes tended to become. They did, however, re-
spond to the glorification of the main figures. 'Ali became a major hero and
Muhammad was given metaphysical status, while Husayn has been bewailed
by many Sunnis almost as much as by Shi'is.

Ismd'Ui piety: esotericism and hierarchy

Quite a different turn was taken by others of the followers of Ja'far al-Sadiq.
Recognizing the line of his son Isma'il as the true imams, they created a

MUSLIM PERSONAL PIETY 379

dynamic social and intellectual movement which fostered numerous rebel-
lions and finally seized power in Egypt in the tenth century, on the decline of
'Abbasi power. Esoteric Shi'ism played a more far-reaching role among the
Isma'ilis (adherents of Ja'far's son Isma'il) than among most Twelvers.

Indeed, they may be regarded as the most successful section of a move-
ment which we may call the Bdtiniyyah, those who gave primacy to an 'inner
meaning', a bdtin, in all religious words and formulations. (Some sections of
the Batiniyyah owned the Twelver imams, however.) Growing out of the
Ghulat theorists, such groups seem all to have been Shi'i in tendency. Their
piety was built on a sense of the esoteric hiddenness of truth and holiness.
These were concealed from the masses, who were held capable only of the
husks of faith, not its inner kernel. It was to 'All alone, the family confidant of
Muhammad, that the inward, secret meaning of the Qur'an had been con-
fided; and only those spiritually alert enough to recognize 'All's position were
vouchsafed such truths, for which they alone were ready. 6

There is something in the Batini mood that resembles that of the Mani-
cheans; in fact, it is probable that some of the same sort of men who at the
beginning of the 'Abbasi regime secretly adopted Manicheanism, a century
later were adopting Batinism and especially Isma'ilism (with equal secrecy).
Both Manicheanism and Isma'ilism proposed to give their initiates a wisdom
and a cosmic dignity which the coarse minds of ordinary mortals could scarcely
aspire to; outsiders were scarcely granted full human status. Like Manichean-
ism, Isma'ilism cultivated its own comprehensive body of science; this was
based on that of the followers of Hellenic philosophy, but was modified in
terms of an esoteric vision of the cosmos as a symbolic whole.

Isma'ilism contrasted with Manicheanism, however (and herein showed its
Muslim character) in being oriented strongly to the practical development of
the world's social order, to the movements of human history. Significantly,
most Isma'ilis, like the other Piety-minded groups, recognized the binding
force of the Shari'ah — regarded as the prime overt work of Muhammad. Its

' In studying the Batini movement, more even than in most other religious studies,
we need to make use of what Massignon calls the psychosociological 'science of compas-
sion'. The scholarly observer must render the mental and practical behaviour of a group
into terms available in his own mental resources, which should remain personally felt
even while informed with a breadth of reference which will allow other educated persons
to make sense of them. But this must not be to substitute his own and his readers' con-
ventions for the original, but to broaden his own perspective so that it can make a place
for the other. Concretely, he must never be satisfied to cease asking 'but why?' until he
has driven his understanding to the point where he has an immediate human grasp of
what a given position meant, such that every nuance in the data is accounted for and
withal, given the total of presuppositions and circumstances, he could feel himself doing
the same. Such a grasp is to be checked, of course; for instance, by testing whether cir-
cumstances which must be presumed, so as to account for an attitude, can then be attested
independently. Yet however risky the method is, it is less risky than any more external
method. Massignon refers to the work of Wilhelm Dilthey and of Carl Jung to illumine
his point of view. Cf. Louis Massignon, 'Les Nusayris', in V Elaboration de Vlslam,
ed. Claude Cahen (Presses universitaires de France, 1961), pp. 109-14.

380 THE CIVILIZATION OF THE HIGH CALIPHATE

high position was symbolized in the primacy of Muhammad himself in
the Isma'ilis' symbolic historical hierarchy — whereas some other Batinis
went so far as to exalt 'All and his secret knowledge above Muhammad.
Yet the Isma'ilis' Shar'ism was not merely reinforced by, but to a degree
transcended by, their 'Alid loyalism, interpreted as the basis for esoteric
truth.

The purpose of the believer was to fulfill himself through the fulfillment of
God's self-realization in the world — that is, of God's fulfillment here of His
own rational cosmic possibilities. The world was no mere indiscriminate test-
ing ground into which souls were placed to see how well they would do their
duty; still less was it a work of blind evil; it presented, in all its details, a
complex and beautiful divine plan. The Isma'ilis — in a tradition which went
back at least to the Pythagoreans — loved to present their sense of an invisible
underlying cosmic order in terms of numerical parallelisms: the seven openings
in the human head answered to the seven visible planets, for instance, and to
the seven days in each quarter of a lunar cycle, and to the seven intervals of a
musical octave. This interest in numbers, though it sometimes took the subtler
form of an interest in proportions, had in common with modern physical
science only a very general trait, the expectation that rational, non-sensory
uniformities were to be discovered, in whose light the most diverse pheno-
mena would prove to have a common nature and a common meaning. The
overriding interest was in finding a physical and moral unity in the cosmos
and in its history, which would invest all details of an individual's life with
cosmic meaning. The natural test of a religious system, then, was the degree
to which it reflected the cosmic harmonies, even in its details, and allowed its
adepts to reflect them through participating in it.

The cosmos itself was conceived in the traditional Irano-Semitic lore in
which the Greek philosophical tradition played a large role, as hierarchically
structured; the Creator was at the peak, working through the diverse circular
motions of the heavenly spheres to evoke all the complex movements of our
sublunar earthly sphere. Corresponding to this natural hierarchy of Creator
and angelic heavens and ordinary mundane life was, for Isma'ilis, a religious
hierarchy. The Prophet was at its head, of course; and the hierarchy was
formed by delegation of authority from him. His single representative in each
generation was the imam of the time, designated by nass in the family of 'Ali;
in him was invested the sacred 'ilm which knew the divine plan and could be
unfolded to those who were worthy. But delegation did not end with the imam.
Indeed during most of High 'Abbasi times, in contrast to the group that was
to become the Twelvers, the Isma'ilis had no accessible imam. Their imam
was 'hidden' since Isma'il's death — his son Muhammad had travelled off, it
seemed, not even the initiates knew where. Instead of a single spokesman, the
Isma'ilis acknowledged twelve chiefs, each with his own territory, to repre-
sent him; and the faithful under their command were further ranked in
various increasingly numerous levels of hierarchy from the Mis, summoners

MUSLIM PERSONAL PIETY 381

or missionaries, down to the simple converts. 7 Those of higher ranks taught
those of lower ranks as much as they were ready to learn of the imam's secret
'ilm.

This hierarchical organization was protected by an extreme use of an old
principle. First those of the Kharijis who did not actually go out in military
bands, and then many of the Shi'is who also had to accommodate themselves
to authority which they could not in conscience accept, had developed the
notion of taqiyyah, pious dissimulation of one's true opinions. It was not only
to protect oneself but also to protect the community of which one was a
member that a Shi'i was urged to practice taqiyyah dissimulation over against
Sunni majorities or Sunni governments: at the least, not to press on their
attention the Shi'i belief that the established Islam and the established govern-
ment were illegitimate and should, in principle, be overthrown in the name of
the imam. Taqiyyah came readily to include not making public among ene-
mies those of the group's doctrines that would be most subject to misunder-
standing. Among the Isma'ilis, now, it took on a more extensive implication:
it became the protecting of the sacred lore from profane ears — even from the
less fully initiated of the adherents; eventually, it was still maintained even
under an Isma'ili government. Taqiyyah dissimulation became the internal
discipline which supported the hierarchy: the lore was protected at every level
from those not yet ready for it.

Isma'ili cyclicism

But the Isma'ili hierarchical discipline was designed not only to guarantee the
soundness of the secret lore; it was appropriate also to conspiratorial political
purposes. For the Isma'ilis expected, far more actively than the Twelvers with
their quiescent imams, a new dispensation for the world as a whole. A dramatic
element was restored to the Muslim sense of history among the Isma'ilis, not
however, in the sense of a nature cycle but on a strictly moral level.

A cyclical sense of history is very natural once the historical process (under
conditions of literacy and urban life) is moving fast and steadily enough for the
individual to be aware of it as a long-term process of change. If an infinity of
space and time are assumed, but a finite range of formal possibilities, then a
certain reflection, applied to the 'old man's sense of time' (the feeling that
'the younger generation is going to the dogs', extrapolated to include the
implication that ever since some idealized starting point each generation has
gotten, on the whole, worse), leads directly to an expectation of cycles in
which renewal is followed by a steady decline. A somewhat different cyclicism
can result from the same supposition of infinity applied to the 'young man's
sense of time' (that the older generation are 'old fogies'), as we see in the

7 On the hidden, 'mastur', imam see the basic studies by Wilferd Madelung, 'Fati-
miden und Bahrainqarmaten' and 'Das Imamat in der friihen ismailitischen Lehre',
Der Islam, 36 (190), 34-88, and 37 (1961), 43-135-

382 THE CIVILIZATION OF THE HIGH CALIPHATE

presuppositions of many modern cliches, which assume patterns of progress
that would recur not only in other societies but even in future species if ours
should destroy itself, or on other planets until some final cooling down. The
fidelity of the religious traditions of Irano-Semitic heritage to the sense of a
once-for-all linearity, which is strong even in Isma'ilism, bears witness to the
high prestige among them of the moral insight that particular actions are
absolutely decisive and not to be reversed or written off as mere incidents in a
recurrent pattern.

The Isma'ili cyclicism, like its esotericism,- represented a revival of view-
points which had generally accompanied in the Irano-Semitic traditions —
perhaps as inevitable logical complement — the doctrine of a single irreversible
historical sequence, with beginning and end, which tended to dominate those
traditions. Elaborately as the world reflected, in some Isma'ili thinking at
least, a divine pattern, it was not free of corruption. As in all the systems of
Irano-Semitic monotheism, there had been a point of initial error and sin.
The Isma'ilis (in the manner of the Gnostics) interpreted the rebellion of
Iblis (Satan), laconically sketched in the Qur'an, as a cosmic turning point,
necessitating an elaborate procedure of restoration, which constituted human
history. (In some Isma'ili philosophy, this initial aberration was identified
with the false sense of independence from overriding cosmic rationality, which
the vital will can be tempted to indulge in.) The greatest of the prophets
(Adam, Noah, Abraham, Moses, Jesus, and Muhammad made up the usual
list) each had an 'executor' who taught the secret meaning of the prophets'
outward revelation and initiated a sequence of imams who in turn remained
unrecognized except by the elite. These missions were to culminate in that of
the Mahdi to come, who would form a seventh in the line and bring in the
purifying Resurrection. 8 (Sometimes this was distinguished from the indivi-
dual resurrection of each of the faithful, which was made an essentially spiri-
tual matter based on transformation of the personality through the truth,
while the great historical resurrection was to include the political establish-
ment of truth and justice in the world.)

The sort of cyclicism worked out by the Isma'ilis can be traced in much of
the later esoteric Islam. However, the cyclicism of early Isma'ili thinking can
be exaggerated. For the most part, the periodicity of the seven great public
figures (ndtiqs), and of the seven imams for each, was emotionally and logically
subordinate to the linearity of moral progression from the initial cosmic aber-
ration to the restoration of cosmic harmony. Their readiness to adopt a politi-
cal role demonstrates the history-mindedness of the Isma'ilis, who were as
convinced as any other Muslims of the special mission of Islam. Later, as the
Isma'ili doctrine became more highly developed, its philosophers interpreted

8 Henry Corbin, in 'Le temps cyclique . . .', Eranos Jahrbuch, 20 (1951), 149-217,
and elsewhere, is the latest to stress the importance of cyclicism in Isma'ili thinking.
The work on cyclicism by Mircea Eliade is also relevant here, notably 'Mythologies of
Memory and Forgetting', History of Religions, 2 (1963), 329-44.

MUSLIM PERSONAL PIETY 383

it by way of neo-Platonism and thus imposed on it an atemporal quality little
in keeping with its historical role. Moreover, as it became involved in public
political events without the expected final consummation, the historical theory
became steadily more refined and complex, incorporating explanations for all
the contingencies which in fact had arisen; in this process, also, its cyclicism
was sometimes carried to the point of overshadowing its sense of linearity.
(This happened at least in the much later Nizari Isma'ili thinking.) But the
Isma'ili core was chiliastic.

Counting from Hasan (for 'All was not a mere imam but the Prophet's
executor), the seventh imam was Muhammad b. Ismail; as seventh, many
expected him to be the Mahdi; he was to take power as soon as the effective
organization of his followers was sufficient. The hierarchical authority and the
esoteric secrecy thus served the individual not only in his personal spiritual
development but at the same time through his participation in an ongoing
historical cause; this social programme was as essential to the divine self-
realization as was the personal development of the elite. (In the tenth century,
in fact, an imam did appear publicly — but only as a descendant of Muham-
mad b. Isma'il — and launched a campaign which first rallied the Isma'ili
movement to the imam and then hoped to win the whole of the Dar al-Islam.)'

During High 'Abbasi times, Isma'ilism came to be the chief vehicle of the
more esoteric of the chiliastic hopes that had gathered around Shi'ism, though
there continued other esoterically oriented Batini groups (notably that which
separated from the Twelvers only at the end of the tenth century to become
the Nusayris, of whom a body still survives in a peasant area in northern
Syria). Isma'ilism offered much to many. To the speculatively inclined, it
offered a remarkably well worked-out picture of the cosmos; in particular,
the rich mythical symbolism which had found a place in the earlier parts of
the Qur'an, but which the moralistic temper of many of the 'ulama.' scholars
tended to reduce to rationalized prosaic irrelevance, was allowed its own vita-
lity. To the intellectual, the shelter of its esotericism provided a worthy place
within an Islamic framework for many interests which in the ordinary courtly
order could well be dabbled in by Muslims, but scarcely as Muslims: philo-
sophy and even natural science. It was under Batini, and probably Isma'ili,
patronage that the most popular of the earlier compendia of Hellenic-type

' Bernard Lewis, The Origins of Isma'ilism (Cambridge, 1940) is still the standard
account of the rise of an Isma'ili imam to power. He may be over-enthusiastic in tracing
the movement to a fairly explicit economic class orientation. His picture of the wider
Isma'ili movement of the time is to be corrected by the studies by Wilferd Madelung
mentioned above; and by articles of S. M. Stern, 'Isma'ilis and Qarmatians', in
L'Elaboration de I' Islam, ed. Claude Cahen (Paris, 1961), pp. 99-108; supported by
'Heterodox Isma'i-lism at the time of al-Mu'izz', Bulletin of the School of Oriental and
African Studies, 17 (1955), 10-33; an d 'Abu'l-Qasim al-Bustl and His Refutation of
Isma'ilism', Journal of the Royal Asiatic Society, 1961, pp. 14-35. These writers have
used the Isma'ili materials to get past not only the tendentious image of the Isma'ilis
presented by their enemies, which used to be reproduced by scholars, but also the
ex post facto image of earlier Isma'ilism presented by the later Isma'ilis themselves.

384 THE CIVILIZATION OF THE HIGH CALIPHATE

lore and science, the Epistles of the Pure Brethren (Ikhwan al-Safa'), was com-
posed (of which more later); and the chemical work of Jabir b. Hayyan —
foundation of the major corpus of early Islamicate chemical studies — likewise
breathes this spirit. Finally, to the concerned man in the busy cities, trying to
get a fairer share of prosperity, it offered hope of social justice and a sense of
active participation in the struggle for this under the blessing of God. Wherever
Shi'ism was found, and sometimes elsewhere too, Isma'ilism was potentially
influential in favourable circumstances. Nevertheless, the Isma'ilis, like the
Manicheans, seem nowhere to have established their allegiance as commanding
the daily faith of a whole population or even of a normal cross-section of it.
Isma'ilism remained the faith of an elite.

Jama' ism: the piety of solidarity

While many Shi'is cultivated a sense of isolation and suffering in a blind and
wicked world from which they awaited a foreordained historical deliverance,
certain of the Jama'i-Sunnis were developing an ethos more appropriate to a
historically successful majority. While the Shi'is stressed the historic mission of
the Muslim community and bewailed itsbetrayal, these Jama'i-Sunnis stressed
its great heritage and cultivated a pride in identifying themselves with it.

In the Jama'i fold, as in the Shi'i, several sorts of piety were in rivalry. Of
the several factions of the Piety-minded opposition which then accepted the
'Abbasids, the Mu'tazills survived the longest as an independent school, while
most others were gradually absorbed in what we shall call the Hadith folk.
They were as active at Baghdad and elsewhere as at Basrah where they had
originated. In Marwani times, the Mu'tazili scholars had sometimes attempted
to find irenic positions, on which all the opposition factions might agree: they
refused to judge between 'Ali and his opponents, and in the disputes as to the
status of a sinner, they chose simply to use the Qur'anic term for 'sinner'. But
then they pursued their chosen positions with logical rigour; thus they refused
to allow the sinner any other status than what was implied in that Qur'anic
term — hence he was neither faithful nor infidel. Under the 'Abbasids, they
were relatively content with the orientation of Islam as it was or had been, and
did not care for either emotional involvement or esoteric lore; their interest
was in moral and especially doctrinal purity.

The Mu'tazilis were noted for the rigour of their personal lives; but still
more for their doctrinal speculation — if they allowed themselves a religious
delight, it seems often to have lain in the charms of logical analysis. They
commonly took a great interest in law, but they supplemented it with a
strongly intellectual concern with ultimate questions. Here they insisted
especially on sound monotheistic propriety. They stressed the responsibility
of free men before a just God: humans' evil deeds must not be ascribed to
God, but to the human beings themselves, who will be justly punished for
them. It was equally important that no weakness proper to creatures be

MUSLIM PERSONAL PIETY 385

ascribed to God — any Qur'anic references to physical or personal traits in God
must be understood metaphorically. Withal, they were sufficiently concerned
with the practical moral commitment to which the Ummah was bound, so
that among the five headings into which they divided their general treatises of
doctrine, along with God's unity, God's justice, sin, and the state of the sinner,
came the duty of 'commanding the right and forbidding the wrong' by tongue,
hand, and sword. 15

The Mu'tazilis were among the first Muslims to push strongly a point of
view which was already represented in the Qur'an, though it did not have so
exclusive a place there: that belief, in the sense of acknowledgement of certain
propositions, was crucial to salvation. The monotheistic conception of faith
implies, initially, an act of will more than one of intellect: at once trusting
God and being faithful to what He requires of one. But in all the monotheistic
traditions, the notion of trusting God — not a convenient basis for identifying
adherents — has tended to be replaced with the notion of believing that God is
trustworthy; the notion of being faithful to His commands has been re-
placed with that of believing that one should be faithful. The notion of belief
allowed a reasonably objective criterion of community allegiance. Hence the
very words that conveyed the more voluntary notion of 'faith' have come to be
understood as meaning 'belief. (Such a rendering of the words makes non-
sense of many passages in the Qur'an, though it fits some fairly well.) The
Mu'tazilis emphasized works as well as belief, but it was insistence on intel-
lectual belief that led them to develop their elaborate systems of doctrine,
defining and defending the proper belief. This point of view was never lost
among Muslims afterwards.

In defending their moralistic logical theses, the Mu'tazilis became the chief
early exponents of apologetic theology, kalam; most later Islamic religious
analysis goes back to them, and we will meet with them further under that
head. They appealed little to the masses, though they stressed the responsible
equality of believers and usually had little patience with aristocratic luxuries.
Their outlook, in fact, seems to have been better suited to a small ruling elite
whose members could be required to stand on their dignity in the face of a
rabble of whom a high faith could not be expected. This the Islam of the
ninth century ce no longer was, if it ever might have been.

The Mu'tazilis were activists. Many of them actively supported the 'Ab-
basi dynasty after it came to power. Some of these evidently hoped to use their
influence with the 'Abbasid caliphs to enforce their rigorous outlook on Mus-
lims generally: first against the Manicheans; later, against those Muslim
'ulama' whose beliefs, they held, failed to maintain, clearly enough, human
responsibility before God — especially, as it turned out, the Hadith folk, the
representatives of that form of piety which ultimately came to be specially

10 The combination of intellectualism and moralism in the Mu'tazilis is nicely set forth
by Isma'il R. al-Faruqi, 'The Self in Mu'tazilah Thought', International Philosophical
Quarterly, 6 (1966), 366-88.

386 THE CIVILIZATION OF THE HIGH CALIPHATE

associated with the Shari'ah-minded Jama'i-Sunni position. Before the end of
the High Caliphate, Jama'i-Sunni Mu'tazilism went into decline in most
places, pursued by the hatred of its triumphing adversaries. (Its intellectual
principles, however, were maintained and elaborated by other groups of a
somewhat different cast of piety, and eventually, in a modified form, came to
prevail generally.)

On the other hand, many Mu'tazilis proved pro-'Alid, continuing the pos-
ture of the Piety-minded opposition. Some of them won disciples in the mili-
tant faction among the Shi'is which eventually formed into the Zaydi sect.
Without the Mu'tazili name, much of the Mu'tazili piety thus prevailed in
Zaydi Shi'ism (and much of its doctrines in other Shi'i groups as well) . Other-
wise, Mu'tazilism as a Jama'i piety gradually faded away.

The Hadithfolk

Far the most influential form of Jama'i piety, by the end of High Caliphal
times, was that associated with the Hadithfolk, the AM al-Hadith, a group for
whom hadith reports about the Prophet formed the chief source of religious
authority. (They are sometimes regarded as the 'orthodox' par excellence.)
They combined with a keen concern for conservation of what had been
achieved a moral rigorism more emotional than intellectual, which led them into
an opposition to the actual current conditions among Muslims, in the name of
an ideal past. Their triumphant yet populistic piety won them a large popular
following, notably at Baghdad.

This tradition of piety seems to have grown up in circles that were looking
back to the jama'ah solidarity of Marwani times — either, as at Wasit (the
Syrian capital in the Iraq), because they had never approved the intervening
revolutions, or because they shared in the feelings of reaction against them
that were already having an effect under al-Mansur. The Hadith folk made a
point of revering Mu'awiyah as an associate of Muhammad, and at first were
inclined to depreciate 'Ali; they adopted the term jama'ah as a favourite
designation for themselves, along with sunnah and hadith. They respected
especially the school of Ibn-'Umar of Medina (which had been acceptable
among the Marwanis), but rallied to themselves many of the heirs of certain
other schools also — many disciples of Hasan al-Basri and of the moderate
'Alid-loyalist hadith transmitters of Kufah (on this account, the worth of
'Ali was recognized, eventually in a compromise formula: as the fourth in
dignity of the caliphs — according to their temporal order — thus still coming
after 'Uthman but having decisive precedence over Mu'awiyah). Thus they
amalgamated several traditions into a new dialogue in which everyone gave
or took a little. This new tradition became increasingly the typical vehicle of
piety among the Jama'i-Sunni 'ulama' scholars, and gained great popular
respect.

Its tendencies toward anti-revolutionary reaction had extensive effects.

MUSLIM PERSONAL PIETY 387

The whole movement of the Piety-minded opposition to the Marwanis had a
certain disrepute among the Hadith folk, and it was they who blackened the
names of most factions in it (even when they respected individuals who had
supported those factions). One of the principles that tended to be sacrificed
in the new 'Abbasi circumstances was that of the duty of individuals to 'com-
mand the right and forbid the wrong'. The Hadith folk did not deny this
duty, especially as among private individuals, but they did not insist on it in
public matters, they recognized the established Muslim ruler on principle.
Nevertheless, in their own way they were oppositional enough, and certainly
not socially passive. It was in their circles that the work of formulating a
Shar'i Islamic order proceeded most effectively: they had won the allegiance,
not at first of the Hanafis at Kufah to be sure (who had been Murji'is), but of
the circles around Malik in Medina, and al-Shafi'i. The Hadith folk organized
themselves informally in Baghdad, with honoured leaders whose word was law
to their disciples. When al-Rashid came to the throne (786), the Mu'tazilis
lost favour and the Hadith folk were encouraged by the regime. Thereupon
they used their growing popularity in Baghdad to launch a social persecution
of dissidents; and on occasion the government even imprisoned some of their
opponents.

The Shar'ism in the piety of the Hadith folk was based on broad community
loyalism which expressed itself in devotion to hadith reports as transmitted
by (or in the name of) all Muhammad's associates — as embodying the wisdom
and glory of the community as a whole. They declared that all the associates
of Muhammad were to be revered on a common basis, and that the explana-
tion of the disputes and enmities which had divided them was to be left to
God. The conquering community of Muhammad carried in itself and in its
unity a special blessing from God, which should not be disturbed by over-nice
questioning. The Twelver Shi'is had come to see truth and justice forever a
losing cause, to see the righteous persecuted and coming to God only through
oppression and suffering in this world. In contrast, the Hadith folk saw the
truth as normally dominant among mankind, and indeed found a test of
truth in the common opinion of the ruling community, in ijma.'; when the
truth was, on occasion, persecuted (as for a time, later, by the Mu'tazilis),
this was an anomalous, temporary storm to be ridden out defiantly. They were
not given to weeping but to a sober sense of responsibility for a world at their
feet. Their piety served as focus for a broad Shari'ah-minded programme of
social order looking back to the homogeneity of the primitive Muslim com-
munity.

This piety, however, was more than community consciousness or a broadly-
based legalism. Their concern with the detail of revealed law had a dimension
of highly personal and immediate devotion, as emotional and imaginative in
its own way as the mourning of the Twelver Shi'is. Their assumption, in try-
ing to develop the Shari'ah law on the basis of the historical community as a
whole, was that the Qur'an and the sunnah practice were alone enough to

388 THE CIVILIZATION OF THE HIGH CALIPHATE

allow a community of human beings to achieve a life of unmediated responsi-
bility to God. This could be felt to be so only if God could be found truly
unmediatedly in the Law itself. Hence when the doctrine arose that the Qur'an
was God's eternal speech — not merely His creature as were human beings and
other things — the idea was eagerly seized on.

Such an idea glorified the textualist tendencies which these men necessarily
were led to in basing their system on the hadith texts rather than on either
community tradition — as the older legists had done — or on private reasoning,
as the Mu'tazilis tended to do. Against either traditionalism or rationalism or
indeed the personal loyalism of the Shi'is.the Hadith folk glorified the explicit
words (nass) of texts held to be sacred — above all, of course, the Qur'an.
Accordingly, the piety of the Hadith folk was given its supreme expression in
the doctrine (expressed in hadith) that the Qur'an which they recited was, as
word of God, not merely another of God's creations but somehow an eternal
cosmic entity, something of God Himself; the more ardent of them were willing
to die rather than say that the Qur'an was merely created. These men admit-
ted no images to cloud for them the face of God; they admitted no heady
incense or sacred music, no kindly saints, and no graciously redeeming Saviour.
But they did want God to be palpable — they wanted to see God when they
died, and right now they wanted to hear Him directly. Their imaginations
were set afire by the presence of God Himself, Whose speech, which was not
other than Himself, was on their very tongues when they recited the Qur'an,
in their very hands when they held it reverently.

Such doctrines were as alien to the primitive Medina piety as was the Shi'i
hero worship. The Mu'tazilis, shocked, accused the Hadith folk of derogating
from the unity of God and from his transcendent majesty by their over-
familiar attitudes. The Mu'tazilis felt that to make the Qur'an, as word of
God, 'not Creator but not created either' (in the phrase of the Hadith folk)
was to set it up to be worshipped beside Him. This was obviously contrary to
the whole spirit of the Qur'an, which the Mu'tazilis were dedicated to de-
fending. With the support of the caliphal authority from al-Ma'mun's time on,
they tested the orthodoxy of the Hadith folk by requiring them to admit that
the Qur'an, like everything else, had merely been created by God. The Hadith
folk could not grant this: such an admission would strike at the heart of their
sense of the immediate presence, in the Qur'an, of God Himself challenging
the human soul.

Nevertheless, they had their own way of dwelling on God's overwhelming
greatness, which was as necessary to their piety as to that of the Mu'tazilis.
They pushed further a position which had already been suggested at Marwani
Medina (and at which Hasan al-Basri had been shocked). In their emphasis
on the supremacy of God over all things, they insisted that it was He alone
who created human acts, even a person's evil acts. God was above any human
criteria of good or evil, of just or unjust; all things sprang from Him, and if an
act was to be regarded as unjust it was because He so labelled it, not from any

MUSLIM PERSONAL PIETY 389

inherent nature in it which God was bound to respect. God could not be
bound! The Hadith folk in turn, therefore, accused the Mu'tazilis themselves of
dishonouring God. The Mu'tazilis, in their attempt to rationalize their faith,
asserted the freedom of the human will which would be rewarded necessarily
by God's justice. The Hadith folk felt that this was to insult God's power
doubly: by assuming that God was powerless to be author of acts labelled by
us as evil, that He was forced to be what we human beings call just; and by
ascribing to human beings alone their evil deeds, as if human creatures could
create, like God, deeds or anything else. They cited the Qur'anic verses which
spoke of the need to submit to God or else find oneself willy-nilly in sin, and
condemned as un-Qur'anic a doctrine which seemed to make human beings
their own masters. No, God was the only true actor in the universe and human
beings were merely his momentary creatures, unfit to judge Him just or un-
just, reasonable or unreasonable. When His Word came to them they dare
not treat it as a mere created thing to be disputed about; they must simply
tremble, obey, and be grateful.

The spiritual temptation which faced the Hadith folk was to try to seize
the creative moment, the Qur'anic point of contact with Transcendence, and
to try to hold onto it — by turning it into something which can be held onto,
and which therefore inevitably must lose the dimension of transcendence.
This the Mu'tazilis could see, and accused the Hadith folk of shirk, of idola-
trously associating something else with God. But the Hadith folk were surely
right in retorting that the Mu'tazilis, in their intellectualism, were in danger
of putting reason in the place of God — that is, of reducing the divine mystery
to an abstract formula of belief.

Ibn-Hanbal and Hadithi populism

This textualist piety did not achieve its success without some heroes of its
own to revere: notably the great hadith-reporter and legist, Ahmad Ibn-
Hanbal (780-855). Ibn-Hanbal from his youth dedicated himself to Islamic
learning, listening to all possible hadith-reporters and memorizing prodigious
numbers of reports. He long lived in abject poverty, which was further com-
plicated by scruples against admitting any personal practice which he could
not base on Muhammad's example as transmitted in hadith. His powerful
memory, his piety and generosity (he later forgave his chief persecutor),
and his good judgment and eloquence made him a pre-eminent teacher of
hadith.

Under al-Ma'mun (813-833), the Mu'tazilis were restored to favour, and
under him and his successors they were allowed to persecute, in a sort of
inquisition called the mihnah, the leaders of the Hadith folk (who had been
intolerant enough themselves under al-Rashid). Al-Ma'mun selected Ibn-
Hanbal to make an example of if he would not recant the notion of an un-
created Qur'an, threatening him with death. His steadfast refusal gave heart

390 THE CIVILIZATION OF THE HIGH CALIPHATE

to the others. His life was saved by al-Ma'mun's own timely death, but he
underwent a long imprisonment and was hailed as a martyr to the cause."

Ibn-Hanbal seems to have been a man of humble background and the piety
which he represented appealed to the simpler folk of the cities, especially of
Baghdad. For by now the different sorts of piety often seem to have reflected
social class divisions even more than local history as under the Marwanids.
In particular, the difference between Jama'i-Sunni and Shi'i, which may initi-
ally have answered partly to temperamental differences, had, as it became
hereditary, fallen into socially significant patterns: in certain groups, a parti-
cular allegiance was reinforced, even among those initially neutral, by appeals
of interest. In the countryside many of the landed gentry, as they became
Muslim, adopted the Jama'i-Sunni position which was official at court. In the
cities many of the wealthy merchants and bankers were Shi'is; the merchant
quarter of Baghdad, al-Karkh, was the Shi'l stronghold there. Many of these
must have been of old Mawali families whose traditions went back to Marwani
times.

Conceivably the Shi'ism of this class represented the greater persistence of
such merchants in holding out for more purely Islamic and egalitarian social
ideals against the 'Abbasi compromise, which cannot have seemed any great
blow to the aristocratic gentry. From this class were drawn many of the katibs
and high administrators of the caliphal state, who were accordingly very
often Shi'is even though serving the 'Abbasids. To the extent that general
culture would flourish best among the upper bourgeoisie, it is probably also
no coincidence that a disproportionately large number of the writers, philoso-
phers, and other leaders in the Islamicate culture of the High Caliphate were
Shi'is. There were other elements in Islamdom (for instance, certain cities like
Kufah and Qum) which had adopted Shi'ism, but the bourgeoisie played a
major role in it.

The lower classes in the city in turn, the artisans and servants and also the
common soldiery, seem to have been largely Jama'i-Sunni; perhaps the com-
mon people saw little practical possibility of an egalitarian political order
wide enough to include the lower strata in any effective way, and preferred
(especially in the capital) to identify themselves with the Islamic community
as a whole, membership in which set them above the dhimmis and the uncon-
verted peasants. They would also have had no objection to contradicting the
wealthier bourgeois; the Sunni-Shi'i riots in Baghdad, during which the Karkh

1 ' Walter M. Pattern, Ahmed ibn Hanbal and the Mihna: A Biography of the Imam in-
cluding an A ccount of the Mohammedan Inquisition Called the Mihna, 218—234 A ,H.
(Leiden, 1897), is a useful English compilation of the data about Ibn-Hanbal. Un-
fortunately, it presupposes without examination some assumptions of the time when it
was written — that the Mu'tazilis were 'rationalists' and probably licentious and that the
Hadith folk were 'traditionalists' in the sense of holding to unaltered community tradi-
tion. The consequence is that the book misses all the problems. (In it occur some examples
of the obscurity that can result when the word 'tradition' is used for a hadith report as
well as in the ordinary English sense, and it is not made clear which is which.)

MUSLIM PERSONAL PIETY 39I

quarter was sometimes pillaged, cannot have been motivated exclusively by
theological disagreements. The majority of the population, of upper and lower
class, were willing to accept the 'Abbasid family as caliphs, prosperity having,
in effect, vindicated their claims, however dubious at the beginning; hence
Shi'ism depended on specific situations of discontent and did not become
generalized among the great body of new Muslims after its first major defeats.
As the population of the caliphal state became Muslims, they became Jama'i-
Sunni Muslims.

At any rate, the Hadith folk found a vigorous response among many of
the poorer classes of Baghdad. In the rest of the empire, in turn, religious
expectations tended to be moulded by what was received in the capital;
though in some provinces — such as Khuzistan and Khwarazm — it was the
Mu'tazili scholars who were able to set the public tone of religion. The view-
point of the Hadith folk was launched toward general success with its victory
in Baghdad.

In no other movement did the traits of populism appear more strongly
developed than in that of the Hadith folk. The Hadith folk were often at
odds, indeed, with the qussas, the pious story-tellers who carried forward the
popular legendry from pre-Islamic times (especially Christian and Jewish lore),
or derived new legends by a lively imagination from the imagery of the
Qur'an. But it was among the Hadith folk that at least the more reasonable of
such tales found a responsible defence. For however much the Hadith folk
might in principle reject such things, anthropomorphism in tales of God,
presenting Him in the image of a human being, and the legendry of spectacular
deeds which prophets could achieve at God's hands, served to support a sense
of personal contact with the divine presence in revelation. The Qur'an teemed
with anthropomorphic images and prophetic wonder tales which the Hadith
folk found no reason to tone down so long as they served to exalt the glory of
God and the honour of His prophets.

Whatever could get itself embodied in hadith reports received a hearing,
then, however much it pictured God anthropomorphically. Some years after
Ibn-Hanbal a riot was caused when the scholar Tabari objected to a preacher
in this tradition who dubiously interpreted a Qur'anic verse to mean that
Muhammad would sit on the Throne with God (like Jesus for the Christians).
The mob wanted to lynch the sceptical scholar. The cosmic figures in the
Qur'an received some recognition, too, if less explicitly than among the
Ghulat theorists: Ibn-Hanbal himself made a point of citing such verses in
support of the cosmic status of the Qur'an. The movement was so closely
identified with the common people that their opponents commonly called
the Hadith folk 'populists', Hashwiyyah. 12

12 A. S. Halkin in 'The Hashwiyya', JAOS, 54 (1934), 1-28, has made clear the
identification of the Hashwiyyah as the Hadith folk, but derives the name from the notion
of 'redundant speech' rather than 'vulgar populace', both notions being carried by the
word Jfashw. But the pseudo-Nawbakhti's 'ahl ul-hashwi wa-l-jamhur' ( which he imper-

3Q2 THE CIVILIZATION OF THE HIGH CALIPHATE

Nevertheless, populism was not carried to the point of sheer concession of
all popular notions; it remained limited by what the 'ulama' scholars could
feel was proper for the people. Accordingly, the Hadith folk were careful to
reject any outright anthropomorphism, which they recognized was inconsis-
tent with the Qur'anic notion of divine transcendence. Borrowing a formula
originally used in legal thinking, they noted that phrases in Qur'an and hadith
that seemed to sanction anthropomorphism in God or other unacceptable
notions should be, indeed, accepted literally — but with the reservation 'with-
out asking how'. Thus God might indeed, as the text had it, step down from
his throne; this was not to be allegorized away as the Mu'tazilis would; but at
the same time it must be recognized that humans could not know what it
meant — in any case, it could not be assimilated to a man's stepping down
from a throne. Taken seriously, such an attitude took back any concessions
which seemed to have been made to the imagination.

Eventually, when the successors of the Hadith folk and of the Mu'tazilis
had fully merged into a common Jama'i-Sunni tradition, this attitude be-
came the basis for a general adoption among the 'ulama' scholars of a drily
rational spirit already foreshadowed by the Mu'tazilis, which turned all the
luxuriant cosmic imagery of Qur'an and hadith into common-sense prose. But
by then the popular imaginative piety was already flowing in other channels.

After the reign of al-Mutawakkil, later in the ninth century, the more ex-
treme representatives of this approach, who called themselves Hanbalis, after
Ibn-Hanbal, tended to have it their own way in the streets of Baghdad and to
a lesser degree elsewhere. Those whom they suspected of a different approach
were sometimes subjected to great abuse and even mob attack. The sort of
piety represented by the Hanbalis never succeeded in swaying the whole
community. Throughout classical 'Abbasi times there were a number of rival
schools. Yet in the end that sort of piety enforced its key dogmas and much
of its temper upon almost all Jama'i-Sunnis. Before this could happen, how-
ever, quite a different sort of Islamic piety had come to rival it, a piety which
was to overshadow the piety of the old Hadith folk in the following centuries:
Sufism.

B. MYSTICAL ORIENTATION

Twelver Shi'is, Isma'ilis, Mu'tazilis, and Hadith folk all developed out of the
Piety-minded opposition to the Marwanids and represented, in one form or

fectly cites, clearly means 'men of the people, of the majority' : they are not verbose, a
secondary notion, but populistic. In the course of a generally excellent study he suggests
that the Hadith folk took the 'easy' way of being 'uncommitted'; but they were 'uncom-
mitted' only where texts in the hadith corpus were mutually contradictory or where such
contradiction might be implied (when Muhammad's associates fought among themselves).
The point of such neutrality was neither to take an easy way out nor, as Halkin also sug-
gests, to express a quiescent fideist wisdom; it was to lay the basis for a positive pro-
gramme, to which the Hadith folk were fully committed even to the point of opposing the
regime.

MUSLIM PERSONAL PIETY 393

other, a kerygmatic orientation in which the historical development of the
Islamic Ummah played a major role. From the beginning, other Muslims
were attracted to a more individualistic piety, concerning themselves with
more personal problems, which a pious man met when he tried to deepen and
purify his inward worship. As this sort of piety matured, it became frankly
mystical: it was inspired, above all, by subjective inward awarenesses emerg-
ing as the selfhood matured, and the historical, the political role of the Muslim
Ummah came to play a minimal role in it. This less historically-oriented Mus-
lim movement was called Sufism at Kufah, and this name came to prevail
elsewhere also for many centuries. To some degree, Sufism shared the traits of
Christian mystical movement and developed it further; but, like the other
main forms of Muslim piety, it was unmistakably Islamic. At least occasional-
ly, men who might otherwise have become Christian monks were converted
to Islam in its Sufi form when they felt the call to a more reflective life. In
creativity, Sufism soon left contemporary local Christian movements far be-
hind.

Spiritual athleticism within Islam: the aspirations of the Sufis

The early Sufis looked to disparate early founders, but soon formed a single
movement, which was closely associated with the Hadith folk. {Sufis seem to
have found the intellectualistic Mu'tazilis uncongenial.) In some cases it is
hard to draw a line between what was Sufi mystical self-examination and
what was Hadithi moralism. Several Sufi masters also concerned themselves
closely with hadith; and most Sufis were Jama'i-Sunnis, at any rate. On the
other hand, not all the early Muslim mystics were clearly identifiable with an
explicitly Sufi movement. An important Jama'i-Sunni group, the Karramis,
who converted and taught independently in Khurasan, seem equally close to
both Sufis and Hadith folk, with whom they quarrelled. They were much pre-
occupied with law and hadith reports and even speculative dogma, contri-
buting to later kalam thinking, but at the same time their founder (Ibn-
Karram, d. 869) contributed significantly to later Sufi analysis of inward
experience. But, by and large, the Sufis formed a reasonably homogeneous
group who kept in mutual contact despite being distributed throughout
Islamdom.

In some ways, but not all, the Sufis represented in a Jama'i-Sunni milieu
what Batini piety represented in a Shi'i milieu. Mysticism, difficult by its
inner-personal nature to share or explain publicly, naturally tends to become
esoteric and to lend itself to all forms of esoteric interests. The esoteric side of
Shi'ism was more prominent among Isma'ilis than among Twelvers, but no
Ja'fari Shi'i group was without it. Among Jama'i-Sunnis, however, a his-
torical esotericism of that sort would have been out of place. But Sufism did
provide an esoteric form of piety among them, which allowed those of them
who were so inclined to explore hidden meanings and personal resonances not

394 THE CIVILIZATION OF THE HIGH CALIPHATE

allowed for by the soberly public Shart'ah. Just as the Batiniyyah stressed
the more personal and esoteric aspects of 'Alid loyalism, the Sufis stressed the
more inward and esoteric aspects of the imaginative piety associated with the
hadith movement, till it overshadowed, though it did not replace, concern
with the Shari'ah law itself. Even more than among the Batiniyyah, the
Sufis' starting point was ever the Qur'an, whose inward meanings they ex-
plored, attempting to get behind the surface of the words. Their technique
was less allegorical or symbolical than that of the Batinis and focused instead
on the personal experience that the words seemed to crystallize; even so,
there were points of contact between the two ways of more deeply reading the
Qur'an. u

The Sufi tradition was later, in a developed form, to dominate the whole
inner life of Islam, Sunni and even to a degree Shi'i; but in the high 'Abbasi
period it was a minority movement. From the first generations of Islam there
had been those who tended to emphasize personal purity and freedom from
the temptations and taints of living in the world. While the Islamic tradition
set its head against monastic celibacy, there were many Muslims noted for their
zulid, a pious zeal which practically amounted to asceticism. This tendency
was important among the circles of the Piety-minded; like other movements,
notably Mu'tazilism, it claimed a source in Hasan al-Basri (d. 728), the austere
saint of Basrah in Marwani times.

But with 'Abbasi times such tendencies ceased to be merely ascetic. Much
was learned by the mystically-inclined from Christian monks, and no doubt
from other sources too, of the inward life of the soul that would detach itself
from the world and love God only. Islam under the 'Abbasids came to include
a general cross-section of society and no longer merely a ruling minority; con-
currently, the mysticism which had long been a part of life between Nile and
Oxus, particularly among Christians, began to flower among Muslims. Great
ecstatic saints arose around whom circles of devoted admirers gathered, to
record their words and their exalted experiences and to try to imitate their
abandonment of all things worldly.

The Sufis honoured the Qur'an as enshrining God's message to Muhammad;
but rather than devote themselves to the letter of its words, they hoped in
some measure to repeat in their own lives something of the experiences which
presumably Muhammad must have gone through in receiving the various
portions of the words of God. They normally accepted the Shar'i approach
in its Jama'i-Sunni forms as valid in its own realm. But they tended to call
the Shari'ah law merely external, a matter of mere outward actions only
incidentally relevant to the soul. The Shari'ah and all related notions were

13 Niyazi Berkes, in 'Ethics and Social Practice in Islam', Philosophy East and West,
9 ( I 959). 60-62, provides an excellently subtle statement of the several ethical tendencies
in the Qur'an and within Sufism. He shows that in both Sufism and Falsafah there was a
degree of reaction against ways in which fiqh jurisprudence and kalam disputation had
departed from the spirit of the Qur'an.

MUSLIM PERSONAL PIETY 395

subordinated to an inward life. By searching the inner meaning of the meaning
behind the words of the Qur'an, they aimed to relive the spiritual states out
of which the words had been formed. Their methods were retirement and
meditation, meditation especially of the Qur'an and of the very name of God;
their results were very often an intense single-mindedness, a very pure ethic,
and a total spiritual orientation of their lives which caused those about
them to prize, as more precious than life, even fragments of their saintly
existence.

In Islamicate civilization as a whole, three movements can be singled out as
of especially pervasive formative effect: the militarizing of agrarian-based
political authority; the assertion of Shar'i legal and social relations; and the
rise of Sufi experience and teaching. Nothing in literature or art, in the
sciences, or in economic development so ramifyingly marked the civilization
in its distinctiveness. We must take special pains to understand Sufism here.

The history of religious experience and awareness has been singularly diffi-
cult to trace. It is clearly not reducible to the history of cult or of dogma or
even of religious and sectarian allegiance. We must try to study the sort of
expectations the most advanced people have been able to have at any given
times as to what levels of spiritual perceptiveness could be attained, and what
kinds of moral or numinal responsibility or responsiveness could be reason-
ably looked for. Almost certainly the rhythms of development of these under-
lying levels of spiritual life have been slower than those of the more visible
levels; perhaps more on the order of the rhythms of technological develop-
ment in pre-Modern times, and correspondingly hard to perceive even at
best.

In the development of mystical traditions in particular, the deeper level is
especially difficult to trace because of the notoriously ineffable character of
mystical experience. Mystical experience is as incommunicable to those insen-
sitive to it as is musical experience to those deaf from birth. But, since a
mystical experience is almost always personal to a single individual, even
among those who are in some degree sensitive to it, discussion of particular
instances is at least as difficult as discussion of a musical piece with one who
has not heard the music nor even seen the score. At most such discussion is
not logically analytic, but impressionistic and evocative.

Yet it seems clear that, in the centuries of the High Caliphate and after, a
new dimension was being added to the expectations mystics had of what
mystical experience could lead to. In Christian Byzantium (and, at least a bit
later, in Hindu India) as well as from Nile to Oxus we find in this period a new
sort of literary expression of a high love-mysticism which seems to amount to
more than just literary style. Though no older forms of religious experience
were dropped, it seems that certain postures of the soul, which had doubtless
been reached occasionally before and in combination with other things, were
now being widely expected among relatively ordinary mystics; and ways
were being found to cultivate what was expected. In Islamdom, this develop-

396 THE CIVILIZATION OF THE HIGH CALIPHATE

merit of a high love mysticism was associated with the consolidation of Islam
in its Sufi form. 14

Mystical life as personal discipline

'Mysticism' and 'mystical', in their technical sense, refer, in the first instance,
to inward personal experience, more or less transitory as an event but en-
during in relevance, which is felt to express or to lead to a special authoritative
and normative relation between individual and cosmos; then the words refer
also to the practices and behaviour that accompany or express such experi-
ence, and finally to the statements used to describe or explain it. The noun
'mystic' is applied to an adept of the mystical life. (But the adjective 'mystic'
is commonly used — in contrast to 'mystical' — in a quite different sense, for
anything occult or symbolic or simply romantic.) There may be a mystical
dimension in most serious devotional religious experience; but we do not
usually speak of 'mysticism' except where the inward personal experience is
itself the focus of devotional attention.

It is usual to think of the mystical as simply an extraordinary occasion in
consciousness. The most spectacular of the mystical experiences are marked,
overwhelming states of consciousness, which are usually exceedingly transient
and are as emphatic as an access of rage or as acute infatuation or as drunken
hallucinations; indeed, the classic instances are held to be yet more over-
powering and intense than moments of apoplectic anger or of climactic or-
gasm, yet at the same time much calmer and deeper. These can be referred to
as 'ecstatic' experiences, in that the individual feels as if he were somehow
beyond himself. But mystical experience has a wider range than ecstatic ex-
perience alone. The more striking events, at least as they appear in the classi-
cal Sufi tradition, are but the peaks of a very widespread type of awareness.
Mystics have almost always described a lengthy mystical 'way', leading by
innumerable small steps from the first glimmerings of devout repentance in
the sinner up to the most ecstatic moments of the saint. Most mystical writers
have spent far more time speaking of the everyday virtues of patience, cour-
age, and unselfishness, as they appear in the mystical perspective, than of
ecstasies or even of the cosmic unity these ecstasies seem to bear witness to.
We may refer to this wider range of inward experience and behaviour, in
which ecstasies appear as special cases, as 'everyday' mysticism.

Freud popularized the term 'oceanic consciousness' for an undifferentiated,
more or less ecstatic condition often described, by witnesses who have felt it,
as a sense of oneness with the universe, that is, with the total environment.
Freud suggested at least one way of accounting for the possibility of such a

" The work of Gustave von Grunebaum drew my attention to this as to so many
other fundamental thrusts in Islamicate history. My treatment of mysticism here has
been helped, I hope, by the drastic reactions of Alex Morin (whose comments throughout
the manuscript have been provocative); and by Eugene Gendlin's sympathetic clari-
fications.

MUSLIM PERSONAL PIETY 397

condition developing organically: the psychical groundwork for it may lie in
the infant consciousness, which does not distinguish between self and environ-
ment. A recapturing or reconstituting of that undivided consciousness in
adulthood — -by any of several means — could be the vehicle for new sorts of
awareness. 15

Such analyses are essential. However independent a meaning one attaches
to moral and religious phenomena, one must, so far as possible, identify the
organic processes involved. Only when we understand such processes well can
we speak with critical precision about the meaning and value of any human
phenomena. Thus our increasing understanding of what is happening in
human love and hate has helped us to distinguish degrees of validityin different
kinds of loving. But we still know remarkably little about what goes on in
aesthetic and in mystical consciousness. Unfortunately, this notion of 'oceanic
consciousness', though suggestive, may emphasize too much one kind of sub-
jective state, which is not the whole of mystical experience and practice and
may not even be essential to it. 16

Without leaving Freud, indeed, we may bring out some other aspects of the
experience by noting how objects and events can evoke unconscious associa-
tions which can lend them overwhelming power. This mechanism surely un-
derlies part of the experience of the numinous, as analyzed by Rudolph Otto;
and nowhere is the numinous more evident than in most forms of mystical
experience. That experience may be seen as a heightening and internalizing of
the daunted awe and enchanted fascination with which any religiously sen-
sitive person may respond to those moments that carry the numinous for him.
Such experiences, though part of an inward transformation of consciousness,
need not carry 'oceanic' implications.

But for purposes of understanding the role of mysticism in developing a
civilization, one must see the more ecstatic mystical moments as part of a
moral process which occurs almost universally. In this perspective, a more fruit-
ful vantage point for making sense of the whole range of mystical experience,

13 This is not exactly the way Freud put it. Freud sometimes spoke as if such an ex-
perience, if it did no harm clinically, could at will be reduced to 'nothing but' what lay
organically at its origin, though he was cautious in his own formulations. But that a com-
plex phenomenon can be interpreted in terms of its components and its preconditions
need not reduce it to that level and deprive it of its own meaning on its own level. Such
reduction is indeed a temptation. If one wishes to distinguish, in such phenomena, be-
tween their organic origin or mechanism and their moral meaning, one must speak with a
special care which was perhaps irrelevant in the clinic.

16 The most popular treatise in English on the phenomena of mysticism is Evelyn
UnderhuTs Mysticism (London, 191 1), which restricts itself largely to the West
Christian tradition (which is not entirely typical). William James, The Varieties of
Religious Experience (London, 1903) (which R. C. Zaehner's work cited in note 17 below,
among others, will help to correct) is not yet out of date. But there have been several
important studies of mystical phenomena recently, which are bringing the whole subject
to a level where neither the old smugness of the mystics as an elite among the uncompre-
hending mass, nor the old disdain of the 'tough-minded' for the aberrations of gullible
enthusiasts will hold up.

39^ THE CIVILIZATION OF THE HIGH CALIPHATE

both ecstatic and everyday, is the commonplace experience of arriving at
relative personal clarity: the clarity that comes when the elements of a prob-
lem finally fit together, or when one has shaken away the haziness after waking
suddenly, or, perhaps most especially, when a surge of anger has died down
and one can look at the situation realistically and with a measure of genero-
sity. All these are first approximations to a fuller, but still relatively common-
place, experience of clarity that can come at moments of retirement and re-
collection: moments when one can look on one's own current resentments or
ambitions with some objectivity, and even gain perspective on and some con-
trol over one's life-passions generally; when, for the time being, what one
might call one's neurotic compulsions cease to be compelling. At best, in such
moments (as many have discovered), one can face the loss of anything one has
most desired, and even recognize one's own worst traits, without either anxiety
or self-pity, and can find the courage to try to be the best one can imagine
being.

In such moments one may experience, for the time being, much of what the
Sufis described when they told of the states to which the soul attains along
the mystical Way. Thus when one has oneself under control, one can, for the
moment, cease needing to worry what anyone else will think of one's acts, if
they be right in themselves. An accomplished Sufi was expected to be enduring-
ly emancipated in this way from desire for people's approval. It was said of
the saint Bayazid (Abu-Yazid) Bistami (d. 874) that, as he was on his way
back from the hajj pilgrimage, the crowds at Rayy met him with excessive
adulation. It was the fasting month Ramadan, and he deliberately took out
some bread and ate it in public. Since he was on a trip, this was legal for him;
but the crowds saw only the fast-breaking and abandoned him. Again, one
can for a moment freely and honestly feel generous and compassionate im-
pulses for those who most stand in one's way or behave most hatefully.
Such an experience helps make intelligible the more enduring and ingrained
compassion Sufis ascribed to Jesus (pictured as the ideal Sufi) when they
told how, as he passed along a road, people insulted him and he responded
with blessings and helpfulness; and when asked why, he answered: a man can
bring forth only what is within him.

Again, a person may find, at a moment of recollection, that he can depend
on some healthy sense of fitness to guide his judgment among the different
possible courses that present themselves, so that for the moment he seems free
of the confusing effects of self-importance or of wilful preconceptions. Such
an experience anticipates one aspect of what the Sufis meant by reliance on
God. Among the Sufis it was a decisive achievement to reach a level of con-
sciousness where one could let one's whole life be guided by the immediate
will of God, and put one's trust in nothing lesser; so that one made no binding
plans but depended on each moment to provide for its own needs. This meant,
among other things, being willing to pass up any seeming opportunities that
did not bear the marks of divine blessing, however dire the consequences

MUSLIM PERSONAL PIETY 399

might be. The tale is told of a wandering mystic who fell into a pit along a
desert route. He started to call out to his companions for help, but felt a stop
in his mind — he must be patient till God should help him. His companions
went off into the distance and he was left alone. At length came a couple of
other men, but again he was not free to call out; even when they saw the pit
and covered it over, he had to remain quiet. It grew dark; then in the dark
something reached into the side of the pit. Now at last he found himself free
to grab hold and be pulled out, and when he reached the surface he found
that what he had hold of was a lion's paw. That mystic was far along the
Way, and had been called upon, correspondingly, to show extraordinary trust
and fearlessness. The moral is not, as it might seem, that the mystic should
depend exclusively on miracles: no mystic taught this. The tale was told part-
ly, no doubt, to suggest that true servants of God are such great beings that
even lions serve them. But it describes faithfully, at the same time, what
dependence on God meant: not waiting for God in person to pull one out of a
pit, but waiting till such means as happened to offer themselves proved com-
patible with one's sense of God's presence. Such an approach shares, on an
advanced level, a like spirit with the more everyday experience of clear judg-
ment during a moment of recollection.

The experience felt in such moments of retirement and recollection can have
a further dimension which anticipates another side of mystical thinking. In
such moments, as inhibiting fears fall away and one no longer feels the need
to put up pretences and defences against any sort of truth — as one's personal
ambitions suddenly appear petty, and one can find the resources to face any
fact about oneself and to love any being capable of needing love — -it is possible,
even among fairly ordinary persons, for a more universal perspective on life
at large to be felt. At such moments one's standpoint is least self-centred;
that is, most nearly the standpoint that some cosmically objective being
might have who could feel totally all human life and yet act in perfect detach-
ment from any particular one of life's pressures. And this is just the moral
standpoint that emerges (where any does) more intensely from the more
striking mystical experiences.

From everyday moments of recollection, with such selflessness as they
bring, the mystic may rise to ever more intense levels of awareness, which
can take ecstatic forms. Ecstasy, carrying with it both intense euphoria and a
sense of total clarity about reality, can occur in very diverse contexts. But so
far as it has played a role in historical mystical traditions, it has done so as
part of a total self-assessment and self -discipline. It is in such a context that
an ecstatic experience ('oceanic' or otherwise) is most likely to carry with it
the moral standpoint of universality. Always the ground of mystical life, in
this historical sense, is a striving for clarity and sincerity; whatever the level
they have reached, mystics, both Sufis and others, have spoken most persis-
tently in metaphors of Light and Truth. To this sort of clarity, the touchstone
of relevance to everyday life will apply. A primary criterion that mystics have

400 THE CIVILIZATION OF THE HIGH CALIPHATE

used to test the 'genuineness' of an ecstatic mystical experience — that is,
whether it is from God or from the Devil (or possibly consists merely in a
similar, perhaps organically related but abortive, subjective event) — is its
enduring relevance to all dimensions of a person's life, including the everyday.
Hujwiri (d. c. 1077), in his description of the true Sufis of his time (whom he
distinguished from the many deluded or even fraudulent claimants to the
Way), says of one of them, 'I found him to be like a flash of love'. 17

Mystical practice begins in any case, with some sort of retirement and self-
recollection: meditation upon the implications of some important truth;
prayer, setting oneself in dependence before some being of a wider level of
vision ; adoration, acknowledging one's pettiness before the greatness of some-
thing that transcends one's own nature. For the Sufis, the focus of meditation
was of course the Qur'an; the object of adoration was the unique Deity to
which the Qur'an summoned.

Meditation then led to a certain amount of withdrawal, for the mystical
way required the concentration of a person's energies. Ridding oneself of any
dependence on satisfying one's desires or one's disgusts could even require
severe ascetic exercises. To accustom himself to not caring whether he was
physically comfortable or not, one man had himself hung by the heels for
hours at a time; to accustom himself to not caring whether he felt hungry or
not, another would eat but a few grains a day. But Sufi teachers usually
warned against pursuing asceticism for itself — for instance, we would say,
being ascetic out of neurotic guilt feelings. All forms of withdrawal, like the
disciplines of athletes, should serve to establish more complete self-control.

With withdrawal came temptations. Ascetic rigours, combined with moral
liberation, could produce heady delights: ocular visions, auditions, and other
abnormal psychic phenomena; all of which, and even the moments of sheer
exquisite ecstasy, one's teachers warned one to discount, lest they become
simply another pleasure one was attached to. As a good Muslim, the Sufi
must return and fulfill his social duties according to the Shari'ah; hopefully
with increased objectivity and sincerity. But he repeatedly renewed his medi-
tation, seeking nearness to God — to discarding concern with all secondary

17 R. C. Zaehner, in Mysticism Sacred and Profane (Oxford, 1957), has distinguished
(much too schematically) 'nature mysticism', relatively commonplace experience of
identity with the environment — attainable also in drug-poisoning and in schizophrenia;
monistic mysticism, in which the experience is more disciplined and leads to the isolation
of the individual consciousness in seeming transcendence of the natural environment;
and theistic mysticism, in which the experience is likewise disciplined, but develops as a
loving response to a Transcendent presence before which nature and the individual alike
are as nothing; and suggests (unconvincingly) that what sets off the 'theistic' or, as he
calls it, 'supernatural', mysticism from other kinds is its ethical consequences. G. C.
Anamati and Louis Gardet, in La mystique musulmane (Paris, 1961), have likewise
made the distinction between supernatural and natural (monist) mysticism, which they
develop more subtly than Zaehner, but also with less comprehensive a sweep.

Such divisions falsify the integrity of the individual experience, but they do make us
realize that each experience can have a definable structure and that, in the sort of experi-
ences that become historically relevant, that structure has consequences.

MUSLIM PERSONAL PIETY 4OI

preoccupations and to being impelled only by larger and deeper rhythms; or
rather, as they preferred to put it, to seeing and loving only Him.

Mystical life as personal freedom

The results of all this in the individual's life varied, of course, as much as did
the mystical process itself, from person to person. Certain main lines were
common. The mystic tended to have a 'spiritual' orientation; that is, at a cer-
tain point he felt no routine to be binding, no law to be sacred, simply because
it was a routine or a law; if he did follow the routine, it was out of conviction
that it suited his place in the whole cosmos at the moment; but ever new
possibilities of meaning might lead him beyond any customary, beyond even
customary Sufi, ways of looking at things. The favourite Sufi word for God
came to be Haqq, simply 'Right' or 'Truth'. The later Persian poet Jalalud-
din Rumi puts this beautifully {Masnavi, n, 3766 ff .) in the image of the duck
that was hatched and raised by a mother chicken. The duck longs to go out
into the water to swim — the water, especially the ocean, symbolizing the un-
charted Infinite of God. The mother, which is Mother Nature, is frightened of
this water, alien to her, and urges the duckling to stay on land. Each soul
must decide, like the duckling, whether to stay on the safe land of routine
predictabilities, of custom and Shari'ah law, as our Mother and all our friends
plead with us to do; or whether to answer the call of our true inner nature and
launch out into the deep.

With some Sufis, this came to mean that all external rules, notably the
whole Shari'ah, were no longer binding on him who had come to live in the
spirit behind the rules. Most Sufis felt that the Shari'ah was binding on them
as on all other Muslims, if only lest their liberty become a stumbling-block to
weaker brethren. If one loved God, one would eagerly obey His command-
ments. For Sufis, the formal salat worship often became the occasion of intense
devotional fervour. But many agreed that the outer rules were valid only
through their inward, spiritual meaning and purpose.

Hence Sufis increasingly tended to minimize differences among religious be-
liefs. In contrast to the communalist exclusivity of most of the Piety-minded,
they readily tended toward a universalistic viewpoint, looking less to a per-
son's religious allegiance than to his spiritual and moral qualities in whatever
guise they appeared. Eventually, after the High Caliphal Period, it became
almost commonplace for Sufis to argue that even idolaters, who fell down and
worshipped stones, were really worshipping the true God: for they were wor-
shipping the best approximation they could find to the Truth that is at the
heart of all reality, including stones. Such a viewpoint was initiated early.
One Sufi sometimes made his disciples recast the Muslim phrase of witness as
'There is no deity but God and Jesus is His messenger' — a statement tech-
nically valid but psychologically upsetting to anyone bound to a specifically
Muslim allegiance.

402 THE CIVILIZATION OF THE HIGH CALIPHATE

Indeed, many Sufis allotted a specially holy place to Jesus as the prophet
of the inward life, of the gospel of Love. For among Sufis, as among other
mystics, Love of God, and hence tenderness to all His creatures, came to be
seen as the heart of the inward life, just as reverence for God and justice to all
His creatures was the heart of the outward life, of the Shari'ah. There re-
mained no doubt, of course, in most Sufis' minds that the revelation to Mu-
hammad was the greatest and the purest of the revelations. Sometimes it was
put thus: that Moses revealed the majesty of God, and the Law which we must
obey out of respect to Him; Jesus revealed the beauty of God, and the Love
which we must bear Him when we catch a glimpse of His Reality; but
Muhammad came with both Law and Love together, revealing both His
majesty and His beauty.

With their spiritual and universalistic orientation, the Sufis naturally tended
to develop a demandingly 'pure' ethic: that is, highly disinterested principles
of interpersonal action. One of the favourite saints of the Sufis was Rabi'ah of
Basrah, who had died in 801. She was said to have taken a jug of water in one
hand and a fiery torch in the other and run through the streets; when asked
her purpose, she said she was going to Hell to put out its flames and to Paradise
to burn up its gardens so that henceforth people should no longer worship
God from fear of Hell or desire for Paradise, but only from love of God Him-
self. The apocryphal tale is told that one day the saint Hasan al-Basri saw
her at a distance on a desert hill, surrounded by gazelles and other wild beasts,
tame in her presence. As he came closer to her, the animals fled away. He
asked her what it was she had that he did not have. She asked him what he
had eaten that day. 'Only a bit of onion.' What was it cooked in? 'A bit of
animal fat.' There he had his answer.

The spiritual temptation of the Sufis was complementary to that of those
of the Hadith folk who resisted going along the Sufi path. For the Hadith
folk, the danger came from the attempt to capture the unformulable in a
formula, to hold on to God Himself within the words of the Qur'an. In such an
attempt, they risked forgoing the spontaneous responsiveness which never
ceases seeking beyond what it has already found, in favour of a disciplined
responsibility to truth already known: responsibility such as had caused
people to receive and live by the Qur'anic challenge when it was first delivered.
Such responsibility was always necessary to preserve the continuity of com-
mitment in the tradition of any group. But, held to too narrow an exclusivity,
such responsibility could impose a conformity which would preclude any new
understanding, smother the creative dialogue which was equally necessary for
any cultural tradition, and devitalize the very tradition it was meant to serve.
The more venturesome Sufis, on the contrary, were devoted precisely to in-
ward, spontaneous responsiveness: responsiveness to new truth, to new human
possibilities, wherever found, such as had made possible the coming of the
Qur'an in the first place, and which kept alive the continuing dialogue neces-
sary to the vitality of the tradition which it had launched. But this responsive-

MUSLIM PERSONAL PIETY 403

ness, if given too free a rein in undiscriminating universalism, risked under-
mining respect for and commitment to the already formulated rules which in
daily practice made possible the ongoing cultural patterns necessary to human
life in a given community.

This form of the dilemma that can arise between responsibility and respon-
siveness became a determining thread in Muslim history for many centuries.
As seems generally the case, an innocuous 'middle way', hoping to combine a
free responsiveness with sober responsibility, would have been at least as
likely to sacrifice the values on each side, in practice, as to combine them. To
be wholehearted and effective, persons had generally to choose one path or
the other; and in choosing the path, they committed themselves to pre-
occupation with its concerns — and to overlooking, misunderstanding, and
despising other matters of concern. A relatively uninvolved — and impotent —
historian may feel that the inward-minded Sufis and the Shari'ah-minded
Hadith folk were complementary in Islamic spiritual life. In Baghdad and in
the provinces, in the last century of the High Caliphate, such a viewpoint was
rare. As the several forms of Islamic piety became more clearly articulated,
the more Shari'ah-minded came actively to distrust the Sufis and were in-
clined to persecute the less cautious of them for heresy; and the Sufis, though
respectful of the Shari'ah and of the hadith-minded circles from which their
movement had arisen, often privately looked down on the more Shari'ah-
minded 'ulama' scholars as concerned more with the husks than with the
kernel of truth.

On reading mystical texts

The Sufis could not dispense with a verbal presentation and even an intellec-
tual analysis of their experience. The awareness come to through mystical
processes cannot properly be put in words any more than can, say, the aware-
ness of music. One may well draw moral lessons from at least the first levels of
deepening clarity. But the content of the more intense levels, especially at the
point of ecstasy, is strictly ineffable. Nevertheless, if there is to be discipline
of the endeavour, there must be some sort of communication about it so that
the sagacity to be drawn from varying experience can be shared. It is perhaps
improper for the person of small or no mystical experience to read such texts
at all. They were certainly never meant for the general public. In modern
times, however, nothing human retains its sacred inviolability; we feel that
for better or worse all barriers must come down and even the amateur must
taste so far as he can whatever has proved or seemed important to human
beings, so as to have maximum opportunity to appreciate what life may be
about. Hence with hands often profane we peruse the old, secret texts. If we
do so, however, it behoves us to use proper respect and caution.

In particular, I must note that in discussions such as these, I necessarily
make those points on which I am reasonably confident, and I cannot steadily

404 THE CIVILIZATION OF THE HIGH CALIPHATE

keep in perspective the areas that have not become clear to me. Yet the whole
realm of the mystical life is peculiarly resistant to so tidy a treatment as I
may seem to be trying to give it.

In reading mystical texts we must keep distinct three components in the
report of any given experience, which the texts themselves do not normally
distinguish, for among devotees this would have been pointless. We must
distinguish the event itself, the process which the given human organism
went through — for instance, a moment of ecstasy or a visual image. This must
have an organic character, latent in the human structure; it cannot happen to
an ape for instance. Many of the classical events have been more or less re-
produced by drugs; in fact, some of the later Sufis, particularly in the Later
Middle Period, themselves made use of drugs as shortcuts, with some recogni-
tion of the implications of such a procedure. Second, we must distinguish the
formulated meaning which the ecstatic or the visionary assigns to the event,
or which he deduces from it. This is a matter of words in which the event is
tied to a wider life context. In different contexts, the same event may have
different meanings; we must see what context it had for the Sufi. Then finally
we must try to sense the experiential content behind the formulated mean-
ing, what the non-verbal referent of the words was in the total experience of
the Sufi; only then can we begin to know truly the meaning of what he was
speaking about.

One may compare to the mystical event an event such as a fright in which
a person thinks he sees his whole life passing before him, and after which
finds strength to be and do what he somehow had never been able to before.
The event itself is certainly not actually a case of total recall, though it is in-
deed a severe trauma. The image the sufferer gives of the event helps us to see
what role the trauma played for him; but if he says that fear of death and of
Hell moved him to reform, we must still interpret such phrases, which refer
to things no one has personally experienced. Only as we see the event in terms
of its deeper implications for his sense of guilt and of responsibility can we see
what really happened at that moment to have such far-reaching consequences.
Or we can compare someone listening to a Beethoven sonata. There is the
physical-mathematical sonic event; there is the critics' formulation of what
has happened; but we shall not begin to understand until we have a feeling
for the music itself as music.

In a narration recorded by a visitor, Bayazid (Abii-Yazid) Bistami (d. 874)
■ — whose aversion to public adulation at Rayy we have already mentioned —
described one of his most intense experiences: T gazed upon Him with the eye
of truth, and said to Him: "Who is this?" He said: "This is neither I nor other
than I. There is no god but I." Then He changed me out of my identity into
His Selfhood . . . Then I . . . communed with Him with the tongue of His
Grace, saying: "How fares it with me with Thee?" He said: "I am thine
through thee; there is no god but Thou".' This presents a famous state of
consciousness which a number of Sufis found themselves in, the 'reversal of

MUSLIM PERSONAL PIETY 405

roles': the worshipper plays the part of God and God that of the worshipper.
It is that which, later, al-Hallaj is said to have proclaimed succinctly by de-
claring: 'And 'l-haqq, 'I am the Truth', that is, God. Bistami analyzes it more
intricately in the passage from which I have quoted. In each case, the ground
of the statement is a peculiar sort of experience, one in which the personal
identity is put in question and yet at the same time consciousness is extra-
ordinarily intense. 18

Bistami's formulation of the experience shows that it had a far different
implication for his life than just any psychic dissociation, or a confused dream
— such he may also have had and disregarded. The phrases form an episode in
a longer narrative of an inner experience which seemed to him to epitomize
what he was coming to in his whole life: a narrative that also paralleled, in
purely psychological terms, the description in hadith of Muhammad's heavenly
ascent, alluded to in the Qur'an. Bistami had been pursuing a long process of
discipline, in which he had been forcing himself to peel off one self-centred
preoccupation after another. This reduction of concern with his own wants was
identified with an increasingly pure and intense concern with the Islamic
figure of the Creator-god, Who alone defined the valid response to created
things that might be wanted. It was like a concern with the beloved in a love
affair: for it was to please God that he must cease pleasing himself. With the
impetus of ascetic tensions, but in this moral context, God seemed increasingly
vivid; he found himself confronting Him as if nothing existed but his self and
God. At length, the confrontation became overwhelmingly intense, while the
meaning he still attached to the self was vanishingly attenuated. This point
was symbolized in a conversation in which — though there is indeed still con-
versation, confrontation — there is only one party to converse, the other having
lost all identity of its own.

As to the ultimate meaning of the experience in its totality, apart from any
symbolic formulations of it, each reader can respond only as his own experi-
ence has prepared him. Doubtless there was some neurosis present; some
might say that Bistami was trying to press one line of consequences at the
expense of all else that goes to make up a balanced organic life; that this
forced him to a logical dilemma which — given his intense preoccupation —

11 The translation here is taken from A. J. Arberry in the Bulletin of the John Rylands
Library, 29 (1956), 36-37. However, the selection there printed is misleading because of
its arbitrary omissions. R. C. Zaehner has made a more exact as well as a fuller transla-
tion in his book, Hindu and Muslim Mysticism (London, i960), in the course of an
attempt to show (again) that Bistami's ideas were monistic and derived directly from the
Indie tradition of Vedanta; and were in turn the source of later 'monism' within Sufism.
Many have taken up this problem. (Bistami's notion off ana' — the disappearance of the
individual in God — is often, dubiously, identified with the Indie nirvana.) Indeed, Indie
mysticism is likely to have had influence in the Irano-Mediterranean regions in some
earlier periods; it would be surprising if there were no contact in the time of the High
Caliphate. Yet there is no dependable evidence in Bistami's case. The last word seems
still to rest with Massignon in his Origines du lexique technique de la mystique musulmane,
rev. ed. (Paris, 1954), wn0 denies both monism and Indie influence in Bistami.

406 THE CIVILIZATION OF THE HIGH CALIPHATE

became a psychological dilemma too, which expressed itself in a nightmare
hallucination.

Another person may feel that what matters is what was done with the
neurosis; that the line of consequences Bistami was following had a funda-
mental human relevance; and that whether the point it drove him to was
nightmarish or liberating, it is a point whose potentialities all must come to
terms with. Always to ask, Is this that I am bothered about what I really want?
is to invite ever again the answer: No, I seem to want this only because I
really want that other beyond it; till one's ultimate reference, what one really
wants or lives for, becomes keenly focused. The bundle of reactions of every
day that one started with as a self becomes paradoxically irrelevant.

But Bistami's experience cannot be reduced even to that (if that were all,
he could have put it in so many words) ; for Bistami was describing a process
and not just a result. The exact meaning for human beings of the 'reversal of
roles' was debated for many centuries among Sufis, with many varying
answers given; some made it the highest, some a relatively low stage in the
mystical Way; some even regarded it as a false temptation, to be feared and
shunned.

The Sufi analysis of the unconscious

However these things may be interpreted, we are dealing with the sort of
awareness that emerges out of the unconscious levels of personality. The
Sufis were aware of this, and sought to pin down as closely as possible what
they called the inner 'secret' of the heart. Already in the classical 'Abbasi
period, Sufis attempted systematic description and analysis of what they
had found, especially from al-Muhasibi (781-857) on, who introduced (despite
his devotion to the spirit of the Hadith folk) the intellectual methods of the
Mu'tazilis.

The central and classical analysis took the form of a description of the
stages of intensification of mystical consciousness. Two series of levels in con-
sciousness were described. The maqdm was the level one had achieved in one's
training and personal growth: patience, for instance, or faith or certainty.
These levels were commonly ordered in a set sequence, one achievement laying
the groundwork for the next; different analysts differed in describing the
sequence. The Ml was the transitory affective condition experienced from
time to time, regarded as a special grace from God. 19 At the upper reaches,
discussion was likely to be especially complex, as 'unity with God' was ap-
proached, spirally, like a moth circling closer into the flame that must con-
sume it.

All this helped one to recognize his own true condition, especially by re-

" The classical treatment of these is to be found in the Risdlah of al-Qushayrl, analyzed
by R. Hartmann, Al Kuschairis Darstellung des SAfttums (Berlin, 1914), where some of
the vast and frequently overlapping technical vocabulary of Sufi analysis is set forth.

MUSLIM PERSONAL PIETY 407

vealing weaknesses — as well as strengths — one might not have been aware of.
For the Sufis were seeking to ferret out the secret sin as well as to evoke the
secret love (for God) latent in each person. This was the aim, in practice, of
the Sufi master (pir) as well as his disciples. Younger Sufis commonly learned
the Way through a personal relation to an older Sufi as pir, master, who served
as examiner and director of conscience; the disciple was expected to be wholly
devoted to the pir and to express to him fears and hopes which had been un-
conscious. (Sufis paid much attention to the imaged symbolism of fantasies
and dreams; but their notions of interpreting dreams mostly strike us as ill-
founded or even absurdly naive.) Only later came a more private discipline
based on ascesis and self-examination. The analyses which Sufis made of what
they came up with were, then, analyses of the contents of the unconscious
mind. It was an early Sufi maxim that he who truly knows himself knows his
Lord.

The Sufi analysis, however, must not be confused with several sorts of more
or less comparable analysis to be found in modern times. The Sufi was not
attempting a general scientific study of personality, though he might find
partly relevant such attempts at it as were made in his time. Further, unlike
the modern novelists, he was not primarily concerned with the complex in-
dividual personality, but with universal human potentialities in which in-
dividual variations were likely to prove a mere hindrance. Nor was the Sufi
even concerned with therapy for ill personalities, despite analogies to modern
psychotherapy since Freud. Elementary therapy might help a person blocked
in his self-exploration by repressive fears, but the moral and emotional dis-
cipline of the Sufis was primarily intended to develop normal personalities to
abnormal levels. The analysis, then, was intended as an aid to understanding
the psychical states of those upon the Way, and also, incidentally, to making
sense of the human place in the universe. The starting point was naturally
that document through which the challenge had come, to which the whole
effort was a response: the Qur'an; it was, then, in terms largely derived from
the Qur'an that the analysis was made.

The residue of that Sufi activity of which the historian can take cognizance
is its discipline, the symbolism which it evoked, and the analyses made of
those symbols. A few Sufis — and among them, perhaps, Bistami, who wrote
nothing and whose disciples did not form a disciplined body till a century or
more after his death — seem to be notable above all for the effective symbolism
which they produced. But the public greatness of a Sufi did not lie in the
intensity of his experience but in the skill with which he turned it into public
channels either by organizing the discipline of his disciples, or by working out,
in verbal composition, a symbolic presentation of life and its meaning or (most
especially) an analysis of particular moral and psychical states as a guide to
pirs as directors of consciences.

The most important of the public Sufis, in this sense, was Junayd of Bagh-
dad (d. 910). As can be seen from the chart of the Sufi pir masters, he learned

408 THE CIVILIZATION OF THE HIGH CALIPHATE

The Earlier Sufi Masters

728 Death of al- Hasan al-Basri, hadith scholar and ascetic who

preached fear of God
776 Death of Ibrahim b. Adham, figure of later importance to some

Sufis (as first mystic)
801 Death of Rabi'ah, woman mystic exemplifying love of God

857 Death of al-Muhasibi, familiar with Mu'tazill positions,

representative of transition of non-worldly piety from

asceticism to full mysticism; wrote on moral purification

through self-perfection; influential on al-Ghazali
861 Death of Dhu'1-Nun al-Misri, associated with Coptic neo-

Platonism, systematically described the mystical states and

stages, also interested in alchemy
874 Death of Abfi-Yazid al-Bistami, first of the 'drunken' Sufis

898 Death of al-Hakim al-Tirmidhi, biographer of early Sufis, wrote

on themes usually associated with Ghulat and Hellenistic

circles; influential on Ibn-'Arabi
910 Death of al-Junayd, pupil of al-Muhasibi, who developed a

comprehensive system
922 Death of al-Hallaj, pupil of al-Junayd, 'drunken' Sufi executed

for heresy; became martyr for later Sufis

from most of the more prominent masters who had preceded him; in turn,
most of the later masters traced their discipline in the Way, at least in part,
back to his disciples. Junayd had a strong sense of the potential integrity of a
human being as a perfectible whole, envisaged by God from the first. He
studied with care the several particular states of consciousness and what they
revealed; at the same time, he went beyond detail. He saw the whole process
of the mystical life comprehensively as a return, on a new level, to that pre-
sence with the Creator which a person had had in a nuclear way when he was
being created. The primal day when (according to the Qur'an) God drew all
persons forth from Adam's loins and, giving them separate existence, con-
fronted them with His challenge was the beginning of sorrow — sorrow which
became deeper as one realized more fully one's separation from Him. But to
seek to escape the sorrow in superficial delights would be a false solution. The
sorrow must be allowed to be deepened in longing for God. Then, in mystical
consciousness, the individual would fulfill the original purpose of his creation,
which had been veiled in the complexities and temptations of common living.
Junayd approved of the expressions of Bistami; but he felt that Bistami
had not gone quite far enough to fulfill his experience. Bistami's ecstatic
utterances, in which he seemed to identify himself with God, had marked him
as the type of the 'drunken' Sufi, who counted it the highest state to be swept
off one's feet through one's sense of one's own nothingness and God's allness,
and so to utter one knew scarcely what. Junayd maintained that this was a
passing phase: the perfected saint would go beyond it to a state of sobriety in

MUSLIM PERSONAL PIETY 409

which his awareness of God would make him a more complete human being in
complete self-possession.

Aware of the horror which all such talk of awareness of God aroused in the
Shari'ah-minded, who wanted no emotive extravagance to intervene between
a man and his daily duties, Junayd took great pains to avoid any scandal in
his private conversations with and letters to the like-minded. One of his dis-
ciples, al-Hallaj, lost patience with such caution. Convinced that all people
could and should open themselves up to the love of God, he wandered about
the lands of Islamdom preaching among the common people the ideal of an
immediate loving responsiveness to God's presence as being better than any
amount of ritual or other external proprieties. His teaching about human
nature and divine love was subtle as well as poetic — and he put it in poetry —
but it was, above all, rash. Like Bistami, he seemed to justify the 'drunken'
state of responsiveness over the 'sober' state of responsibility. When he began
making disciples in high places in Baghdad, the authorities found reason to
imprison him as a heretic. Eventually he was tried, condemned as having
taught that a symbolic and spiritual fulfillment of the law was as good as the
actual rites, crucified with his hands and feet cut off, quartered, and burned
(922). It is said that in his agony he expressed his delight that he was suffering
so for God's sake, but acknowledged that his judges were as right to condemn
him, so as to safeguard the community life, as he was right to express para-
doxes so as to proclaim the love of God. 20

20 The student of Sufism using Western languages has three sorts of resources. For a
description of Sufi activities and organization in a given modern area, see Octave Depont
and Xavier Coppolani, Les confreries religieuses musulmanes (Algiers, 1897), chiefly on the
Maghrib; John Subhan, Siifism (Lucknow, 1938), on India (uncritical); and the central
chapters of John P. Brown, The Danishes or Oriental Spiritualism (London, 1867), on
the Ottoman lands, the many errors of which are only partially corrected by H. A. Rose's
edition of it (Oxford, 1927), but which is only limitedly replaced by John K. Birge,
The Behtashi Order of Dervishes (London, 1937), excellent; Joseph McPherson The
Moulids of Egypt (Cairo, 1941) is a smugly ignorant eyewitness. On the classical Sufi
works, we may cite the writings of Reynold A. Nicholson, careful and important studies;
of Margaret Smith, sound on the earliest period; of Louis Massignon, whose Lexique
technique de la mystique musulmane (see note 1 8 above) traces with keen documentation the
spiritual development of the early period, and whose Passion d'al-Halldj, martyre mystique
de I'Islam (Paris, 1922) is a masterpiece which also includes much on the later period,
and all of whose studies present a profound spiritual perception of what was going on,
which, however, must be checked as to detail; of Louis Gardet, with reasonable Catholic
views; of R. C. Zaehner, overschematic but not negligible; of Henry Corbin, hard to
follow and pre-committed, but profound and essential; of A. J. Arberry, useful chiefly
for the long translated portions he uses. Of Sufi writers in translation, Kharraz is
systematic, dealing with moral progression; Kalabadhi has many anecdotes but is
defensive in point of Shar'i propriety; Hujwiri is systematic, covering all disputed issues
of practice in a Shari'ah-minded sense; Shabistari presents unitive metaphysics; J ami's
Lavd'ih briefly discusses metaphysical issues; 'Abdulqadir, Fuiuh, is moral and mystical
sermons; Junayd's epistles, on mystical experience and discipline, are paraphrased very
misleadingly in the English; Ibn-al-'Arabi, Fusils, on metaphysics, is abridged even in
the French. Muhammad Ghazali gives intellectual defences; al-Qushayrl is systematic
on spiritual states. We need more translations of MakkS and Junayd, fundamental guides
to Sufi, and of Ibn-al-'Arabi and Yahya. Suhravardi, on metaphysics and visional life.
Poets much translated are Rumi and 'K\\zx, eminently expressive.

Speculation: Flasafah and Kalam
c 7S°-94-S

The way of life and thought of the Shari'ah-minded grew directly out of
those Irano-Semitic cultural traditions which stemmed from and further
developed the prophetic summons of Axial times. These traditions were
expressly transformed with the advent of Islam and recrystallized in new
common patterns. They were sustained by the high seriousness represented
in the various forms of Muslim piety and in turn came to form the most
dependable basis for the legitimation of the social order. The intellectual
disciplines cultivated by the Shari'ah-minded came to be thought of, among
ordinary Muslims, as the pre-eminently Muslim disciplines par excellence,
and as essential mainstays both of social order and justice and of personal
dignity and purity.

But lrano-Semitic culture had also shown another face from Cuneiform
times on: one in which not the moral judgments of history but the rational
harmonies of nature were the source of inspiration. This tradition had its
own high seriousness in life, as alien to courtly elegance or frivolity as was
that of the monotheistic tradition. But Islam as such initially was no part
of the dialogue of this rationalistic tradition and indeed seemed quite alien
to it; at first, it continued its way side by side with Islam, little affected
by it.

The independence of the two dialogues was partly reflected in a linguistic
difference. The Abrahamic monotheistic traditions were associated largely
with various forms of Aramaic (Syriac) and with other confessional languages.
But, as we have noticed, one of the results of the Hellenizing of some sectors
of society from Nile to Oxus after Alexander was that the rationalistic
tradition of mathematical science and associated philosophical thought came
to be pursued in Greek and associated with the Hellenic tradition. In both
the Sasanian and the Roman provinces at the time of the Arab conquests,
this was still largely the case. Scientific works were being increasingly
translated into Syriac and to a lesser degree Pahlavi, but even in Syriac
the tradition was still associated with the Greek masters.

Both the Abrahamic prophetic tradition and the Hellenizing philosophic
and scientific tradition had, in their origins, dealt with comprehensive life-
orientational problems. Even the mathematical and scientific traditions of
Cuneiform times were instrumental to larger religious visions. The transition

410

speculation: falsafah and kalam 411

into the Greek language had at the same time been a transition into a new
religious framework: that of the Socratic tradition of Philosophia, to which
the particular scientific traditions were more or less ancillary. Socrates and
Plato, by the definitions of religion we have been using, were as much religious
figures as Amos and the Isaiahs; geometry or astronomy were almost as
subordinate to the total cosmic vision which adherents of the several Socratic
traditions were working out as was Hebrew historiography to the spiritual
vision of the adherents of the Abrahamic tradition. But in contrast to the
Abrahamic, the Socratic traditions had appealed, from the first, only to an
intellectual elite; they made no effort at forming their own popular religious
patterns. Everyday cult and popular religious ideas were left to the established
priesthoods and to the folk traditions; at most, the Philosophers suggested
limited reforms or reformulations. With the rise of the monotheistic traditions,
especially Christianity, these took the place of popular paganism in the
Philosophic outlook, being accepted in much the same spirit as the former
religion.

However, the advent of the monotheisms made a difference: the exponents
of the monotheistic traditions were more comprehensive in their demands
and jealous of any alternative life-orientational outlook. Philosophia could
not afford to appear in an explicitly religious role, even for an elite. Hence
the exponents of the Socratic or Philosophic tradition tended to emphasize
the autonomous validity of the several specialized elements in their tradition,
as independent scientific traditions in which a person could participate
regardless of his religious allegiance or even his personal spiritual outlook.
Not only their mathematics and astronomy and logic were so treated, but
even their metaphysics was sometimes regarded as an autonomous discipline,
whose axioms implied no spiritual pre-commitments and which could be
pursued as a sheer intellectual exercise — a notion that would have been
as repugnant to Plato, for partly different reasons, as to modern existen-
tialists or linguistic analysts.

This accentuation of the autonomy of the several scientific disciplines,
insofar as they could claim to be independently demonstrative, opened their
dialogues to wider participation, in principle, at the same time that it tended
to limit the sorts of questions that could be pursued within those dialogues.
It also resulted in an ambiguity as to the place of the more life-orientational,
the more religious aspects of the tradition, which had to present itself simply
as 'philosophic' human wisdom, and which came to seem, even to some of its
adepts, to be merely uncommitted rational understanding. Nonetheless, in
practice, the Philosophic tradition as a whole continued a common dialogue
in which those who pursued the several specialized disciplines generally took
part, and which continued to presuppose a more or less common world view,
common spiritual commitments, and even, in large measure, a common
manner of joining the Philosophic spiritual commitments with those of
prevailing popular religion.

412 THE CIVILIZATION OF THE HIGH CALIPHATE

The adherents of the Socratic Philosophic traditions readily formed a
world of their own aloof from any society in which they found themselves.
By High Caliphal times, the tradition had become associated in a practical
way with Christianity, since this was the communal allegiance of most
Syriac-speakers. Christianity itself had been profoundly touched by it:
Christian thinkers had had to confront the Hellenic metaphysical and
logical traditions, and the formulation of the problems of Christian theology —
problems concerning the nature and power of God and the freedom of human
beings — reflected this. At the same time, the Hellenic intellectual traditions
persisted independently of Christianity and were pursued by Christians and
non-Christians alike. And in the Mesopotamian town of Harran, the com-
munity of Hellenistic pagans still cultivated the Hellenic tradition as an
all-sufficient spiritual and cultural whole.

The practical disciplining of the natural-science heritage

Whereas the objective studies proper to the Shari'ah-minded, that could be
pursued in relative autonomy from an Islamic commitment, were especially
historical studies, from the collection of hadith reports to the elaborate
compositions of Tabari, the Philosophic tradition expressed itself most
objectively in nature studies, particularly those based on mathematics.
Without an intensive culture in positive sciences, a rationalistic philosophic
approach to life could be as vacuous as one that attempted to found itself
in the prophetic spirit without a solid historical sense. Moreover, such
positive sciences could be eminently useful. When the learning of regional
rationalistic traditions was being rendered into Arabic, it was naturally first
technical texts in medicine and astronomy that were most translated. Only
later could serious metaphysical work be carried on in Arabic.

From the time of al-Ma'mun through the ninth and well into the tenth
century, translation of older texts, primarily Greek or in the Greek tradition
(Syriac or Pahlavi), but also some from Sanskrit, was the most important
sort of scientific activity. The Christian Hunayn b. Ishaq founded a school
of translators who were themselves of high scholarly standing and were able
to improve greatly on the first translations that had been made as well to
enlarge the range of the translations. They took pains to gather as many
manuscripts of a given work as possible and to collate the texts, so as to
have a sound basis for translation to begin with: a major achievement in
days before printing could avoid the errors of uninterested hired transcribers,
and essential if the texts were to serve as more than bare hints to subsequent
investigators. Thus, gradually, the full Greek philosophic and scientific
heritage became far more freely and accurately available in Arabic than it
had been in any Sasanian language. The results showed themselves in a
steady enrichment of Arabic scientific writing with old Greek materials.

Perhaps almost as respectable an achievement was to work out both a

speculation: falsafah and kalam 413

technical vocabulary and a pattern of syntax which would allow scientific
work to be carried on fluently in Arabic. Every human language is potentially
adaptable to every linguistic use, but when it has not yet been used for
abstract discursive purposes, every language presents its own obstacles
when such use is attempted. Arabic was not rich in means of subordinating
one phrase to the next, for instance, but the translators established patterns
which served all purposes effectively. With this work, the labour begun by
prose-writers such as Ibn al-Muqaffa' was completed, and Arabic became a
written language sufficient to carry the whole range of pre-Modern culture. 1

Within the same period, Arabic was being used for important scientific
advances, largely the work of Muslims. Some of the most notable advances
came only in a later age, but in the last century and a half of the High
Caliphate more discoveries were made than in many centuries previously
either from Nile to Oxus or in the Mediterranean lands. Some of these were
due to the confrontation of the Greek tradition with part of the Sanskrit
tradition — a process which had begun in Sasanian times after the flowering
of Indie science had come to Sasanian attention. Much of the advance was
due to a practical orientation which was brought to scientific studies — as
often happens — when circles take them up who have cut themselves loose
from the older context of those studies and hence from the symbolic over-
tones which had come to surround them; or who have never even been
acquainted with that older context.

Perhaps the most generally appealing of these studies was astronomy.
The earliest of the nature studies to be highly developed almost anywhere, it
yielded dramatic and imaginatively satisfying results to the application of
elementary but precise observation. But the results could be rather too
satisfying. For the Greek tradition, the temptation was great to find in
astronomy just the perfection which their vision of pure reason called for,
in the shape of the universe as a whole.

Aristotle had projected a system in which the seeming movements of the
stars and planets about the earth were reduced to exactly circular movements
in concentric spheres, movements which, moreover, were accounted for
through physical laws. Ptolemy had developed a mathematically more
accurate astronomy designed to account for actually observed positions of
the planets in complete detail. He continued to use circular paths in charting
the geometry of the movements; such circles not only fulfilled the general

1 Every language tends to have its own particular genius, of course, even apart from
the way it happens to develop when cultivated in literature and scholarship. The
spiritual genius of Arabic has been studied imaginatively — if not always persuasively —
by Louis Massignon in several articles. Nevertheless, the common remark — that one
thing the Arabs 'contributed' to Islamicate civilization was their 'amazingly adaptable'
language, which, though hitherto used only by crude Bedouin, was found suitable to
the highest needs of science — bears little weight. There is nothing special in such a
feat. The much-praised Semitic system of vowel-pattern changes for forming verbs
and nouns may be more rhythmically suggestive than a system of prefixes and suffixes;
but it is, if anything, less flexible and less precise.

414 THE CIVILIZATION OF THE HIGH CALIPHATE

assumption that bodies not subject to outside interference must naturally
so move, but also proved easiest to handle in point of mathematical geometry,
which called for regular movements if any calculation of them was to be
made at all. But his were circles within circles, unaccountably complex;
and indeed he renounced any effort to show how the movements were actually
produced, being concerned only to trace planetary paths which would answer
to actual observations, and so to allow maximally exact prediction of
planetary conjunctions and so on. By the time of the Arab conquest,
Aristotle's observationally less satisfying system was still preferred in some
circles because it formed part of a larger system of physics in which, apart
from such details, everything was beautifully accounted for.

As late as the time of the astronomer Muhammad b. Musa al-Khwarizmi
(d. 844 or later) — from whose name we derive the term algorism, on account
of his arithmetical studies — it was the Aristotelian system that was rendered
into Arabic. But under Muslim rule well-equipped observatories were built
and observations were refined still further than in earlier times. (Al-Ma'mun
took great pains to see that a degree of terrestrial latitude be accurately
measured.) By the time of the Jewish astronomer Rabban al-Tabari, the
practical advantages of the Ptolemaic system had become obvious; from
that point on, Islamicate astronomy built upon Ptolemy, as did later the
Occidental astronomy dependent on it. 2

Mathematics, though important in other connections also, was especially
the handmaiden of astronomy. Hence it claimed special attention in the
form of geometry; Arabic- writing mathematicians turned to developing,
especially, spherical trigonometry. But in late Hellenistic and Roman times,
the new mathematical openings were in the realm of algebra. Though we
know little of what happened, it seems clear that the algebraic outlook was
inherited from the later Cuneiform mathematics, whose tradition now came
to the fore as more efficient than a strictly geometrical approach. In any
case, under Muslim rule, algebra was pushed very far along with a revival
of other useful traits of the Cuneiform tradition. The solution of equations
of higher degrees was a fascinating puzzle pieced out bit by bit.

Muhammad al-Khwarazmi brought together much of the Hellenic and
Sanskritic traditions, as they had been present in the Sasanian schools,
including elements of the Cuneiform tradition which have survived otherwise
neither in Greek nor in Sanskrit, in a synthesis which established the import-

2 O. Neugebauer, 'The History of Ancient Astronomy: Problems and Methods',
Journal of Near Eastern Studies, 4 (1945), 1-38, in a general review of the state of studies
in ancient science at that time, brings out well how older views could survive alongside
newer ones in an age when scientific communication was largely a matter of chance (and
how isolated discoveries of importance could disappear without a trace — especially, one
can add, if they were bound up with outdated techniques, such as the complex geometry
used by many Greeks in the place of simpler algebraic methods). It was probably not
only Islam and Arabic but the use of paper that increased scientific communication
somewhat at this time.

speculation: falsafah and kalam 415

ance of the field and defined its most important initial methods. From the
name of his chief book, which, however, was as much a practical handbook
of reckoning as a theoretical exposition, Western languages have derived
the term 'algebra'. In mathematics, one of the most useful techniques,
which he helped to popularize, was the system of place-value numerals with
a zero (instead of the usual alphabetical systems which were nothing but
short notations of the numbers' names). A system of place-value numerals
had been used in the Cuneiform. The Muslim scholars found the system with
zero in the Sanskrit tradition and gradually generalized its use — sometimes
on a sexagesimal base of sixty as well as a decimal base of ten, following
ancient Cuneiform precedent.

Apart from astronomy, mathematics was important in the physical studies,
such as optics and music, the theory of which, already partly developed in
Greek, was eventually given a great impetus in Arabic. The study of musical
proportions was thought of not only as the study of sound (mathematizable
in the form of distinct musical tones in the octave) but as an introduction
to proportion generally: in particular, as a tangible model of the 'music of
the spheres', the system of proportions expected to prevail in the motions
of the heavenly bodies. Music received the attention of the most prominent
scholars.

With medicine, we move into a different circle of nature studies, the
biological, where the use of mathematics was attempted with much less
success. The centre of attention here was human anatomy and its pathology.
Building on the work of the Greeks, the Islamicate physicians maintained
their high traditions (including the Hippocratic oath) and, like the latter-day
Greek and Syriac physicians, tended to be tied to the classical books. How-
ever, many of them were acute clinical observers. The most famous physician
of High Caliphal times, Ibn-Zakariyya.', al-Razl (d. 925 or 934), pinpointed
the distinction between measles and smallpox as well as solving other less
dramatic cases. The caliphal administration attempted to maintain high
standards in the public hospitals and provided for the examination of
physicians, attempting to suppress quacks through the muhtasib officer, at
least in the capital. 3 But medicine, with its very practical use, was a privileged
case; biology at large was not very seriously pursued (indeed nowhere had
it been so) . The practical study of plant and animal breeding and agriculture
generally, in which much advance was made in Muslim times (for instance,
in the development of hybrid citrus fruits still in use) and to which many
manuals were devoted, was kept largely separate from the general body of
natural science.

1 Joseph Schacht and Max Meyerhof, 'The Medico-Philosophical Controversy be-
tween Ibn-Butlan of Baghdad and Ibn-Ridwan of Cairo', Egyptian University, Faculty
of Arts, Publication 13 (Cairo, 1937), a badly written paper, nevertheless gives one a feel
of the scientific life of the time, for instance the broad sense in which a concept like
'heat' was understood for medical purposes, almost analogous to our 'energy'.

416 THE CIVILIZATION OF THE HIGH CALIPHATE

Alchemy

Because of the usefulness of its results to the pharmacopoeia, however,
some contact was maintained between medicine and chemistry, which in
fact dealt largely with organic materials. Al-Razi, in particular, is known
for having developed, from the chemistry of the time, several effective
medical preparations. While astronomy may best symbolize, among the old
sciences, the sheer grandeur of the power of the human mind to comprehend
what at first seem mighty and mysterious forces incommensurate with human
life (its chief application was in astrology, whereby people sought to under-
stand fate), chemistry may best symbolize the human drive to control such
forces and bend them to human ends; a drive almost equally early, if less
pervasive (before recent times) in forming the scientific tradition. At the
heart of chemical studies in both Greek and Muslim times was the practical
application that we call alchemy, the search for ways of transmuting one
substance into another and, in particular, less valuable metals into silver
and gold.

The alchemical tradition seems to have originated in Greek times in the
use, for producing special effects, of recipes developed by ordinary craftsmen
in their work. Scholars were aware of the Aristotelian theory of the two
sorts of quality — fluidity and temperature — which were differentiated as the
four elementary qualities: moist, dry, hot, and cold; which in turn accounted
for the four simple elemental substances of which all compounds were
composed in different proportions: air (hot, moist), fire (hot, dry), water
(cold, moist), earth (cold, dry). On this basis, all ordinary substances were
regarded as compounds which could presumably be taken apart and put
together again differently. Combined with the practical experience that
could be had with the craftsmen's recipes, such a theory led to a conception
of the metals as condensed vapours which in vaporous form could be re-
combined and reconstituted. With an ever-increasing battery of furnaces
and stills, the chemists set about producing what new combinations they
could.

The more philosophical-minded among them seem to have been chiefly
interested in the theoretical implications of their studies. They expected to
learn not merely about the recombination of the elements into various
metals and other substances, but about the analogical processes which they
supposed went on in the universe at large and in the human soul, which
could be illustrated from alchemical experience insofar as all existence is
built on the same principles. The human being was regarded quite literally
as the microcosm, the world in miniature, within which all the cosmic
processes repeat themselves in the organism. A great goal of the alchemist
was to purify the various substances to the highest degree possible, in the
thought that in its truly pure state a substance was incorruptible: the
incorruptibility of gold showed its peculiar chemical purity. But for the

speculation: falsafah and kalAm 417

philosopher, what really mattered was what one could learn from all this
about the purification of the human soul. The more philosophically inclined
alchemists were discussing, in highly developed symbolisms, basic psycho-
logical principles quite as much as they were discussing material chemical
ones. Thus the mutual attraction of man and woman was held to answer the
chemical fusion of contrasting elements. (In consequence, they were some-
times rather careless in exactly assaying the merely material substances they
produced.)

It is doubtless on this account that it was under the auspices of Isma'ili,
or at least Batini, piety that arose the most important corpus of chemical
literature in High Caliphal times : the writings associated with the name of
Jabir (Lat. Geber) b. Hayyan. Jabir was a historical personage, a disciple
of the Shi'i imam Ja'far al-Sadiq (d. 765), but most of the works ascribed
to him — which contain many different sorts of doctrine — are of later dates.

Yet it is in the Jabir corpus that reappears, in the Muslim period, a tendency
which had been less prominent in late Roman times : a strong interest in the
practical as against the symbolical side of the study. The material theory
was further developed: for instance, all metals were held to be reducible to
two primary ones, ideal sulphur (an embodiment of the hot and the dry)
and ideal mercury (an embodiment of the moist and the cold) from which
if they could be gotten in their really pure forms, all other metals could be
developed, including gold. An attempt was even made to work out the
mathematical ratios that would assure purity. Such theoretical development
answered to an increased production, in practice, of elixirs of all sorts (that
is, artificial chemicals used to produce further chemical reactions); some
had explicit medical uses and some, it is said, were used commercially to
make surrogates or, less honourably, unavowed substitutes for various
wares — especially for drugs. Some of the practitioners, doubtless, dishonestly
made the most of a dubious skill, but others certainly believed that they were
producing somewhat impure forms of the genuine article: if their gold, for
instance, tarnished, it was not that it was not gold but that it was not yet
quite pure.

The alchemists, naturally nonconformist souls and readily suspected by
the masses of magic and commerce with the devil or of subversion or at
least of counterfeiting, when not regarded as sheer imposters, naturally
had to do their work as secretly as possible. and often in intellectual isolation.
In the mass of often highly conjectural theory and practice which they
developed, sometimes means were used, such as the numerical relations in
'magic squares' or even in alphabetic symbolism, where the proportions
resulting are purely specious to the eye and can have no deeper grounding,
either theoretical or empirical. And doubtless sheer magic did occasionally
sift in : the use of means whose efficacy is thought to presuppose not intelligible
natural processes, however recondite, but the intervention of special occult
forces, often personal to the practitioner or else thought of as non-human

418 THE CIVILIZATION OF THE HIGH CALIPHATE

beings subject to his control by special techniques. But scholars recognized
the radical difference, in principle, between chemical processes and magic.
Whether or not they thought that the alchemists could in fact succeed in
transmuting substances (or whether, as some did, they maintained that
only magic could achieve this), they recognized that the processes they were
primarily operating with were a genuine part of nature. Chemistry formed a
recognized part, though a part much less honoured than astronomy or
mathematics, of the total body of positive science on which a Philosophic
view of the world could be founded. 4

The Faylasuf s and society : medicine and astrology

As in earlier times, so also in Arabic and among carriers for the most part
Muslim, Philosophic tradition maintained its overall unity, which led its
scholars far beyond a mere academic occupation with the several positive
sciences. One who participated intellectually in the tradition by his studies
and researches was called a 'Faylasuf (the Greek philosophos) ; the tradition
as a whole was called 'Falsafah', philosophia. The word is originally identical
with our 'philosophy', but it implied not just the study of metaphysics and
logic, nor even these disciplines plus the positive sciences, but, more funda-
mentally, the Philosophic approach to living, of which interest in such studies
was an expression.

The Faylasuf ('Philosopher') was dedicated to governing himself by his
reason. Reason, such as we find impurely expressed in our human selves,
was thought to underlie the cosmos as a whole. All aspects of nature were
studied in the spirit of seeking always the uniform logical, rational, principles
behind the apparent diversity of manifestations. This study included the
human being itself, both as body (to be understood in the science of medicine)
and as rational being, to be guided by the science of ethics and politics.
Given this rational knowledge of the first principles of nature and of himself,
everything else followed for the Philosopher, both his personal manner of
life and his conception of how society as a whole should be ordered. The
more nearly a person could approach to such rationality, the more nearly
he could fulfill his own purpose in existence and be at harmony with all
life. For the Faylasuf, the ideal man was a sage 'philosopher' — a man who
fully lived up to the demands of the tradition of Philosophia. (When referring
to 'philosophy' and 'philosophers' in this idealized sense, I shall sometimes
put them into quotation marks to set them off from any special subject
matter; and I shall capitalize references to the actual historical tradition
as a whole.)

* The most useful introduction to Islamicate chemistry is the first half of E. J.
Holmyard, Alchemy (Baltimore, 1957), which stresses the physically chemical side of
the tradition at the expense of the psychologically philosophical side. Holmyard's
historical sense is conventional at best, and very uncritical, but he 'writes clearly and
carefully.

speculation: falsafah and kalam 419

The Faylasufs stood in conscious contrast to the mental world of the
Shar'i 'ulama' scholars, though they often regarded themselves as good
Muslims ; at the same time, they rather looked down upon the sort of society
represented by the luxurious and despotic court. It cannot be said that the
Faylasuf scholars actually represented a total social programme, in the
way that both the 'ulama' and even the men of courtly culture, the adibs,
tended to do so; yet their thought embraced all aspects of living. Some of
them even seem to have had serious hopes of seeing society remodelled
according to their own conceptions. In any case, their attitudes had con-
siderable influence not only among the adibs but eventually among the
'ulama' themselves. As had already begun to be the case in the Sa.sa.nian
empire, Falsafah in High 'Abbasi times came to play a prominent, though
marginal, role in cultural life. But in order to do this, its carriers, the
Faylasufs, had to find for themselves a niche where, despite the contrast
between their tradition and the dominant ones, they could win social
recognition.

Hence the adepts of Falsafah, like the 'ulama.' and the adibs, tended to
follow certain professions. Most typically they became physicians and
astrologers, practicing the two most widely recognized lines of 'applied
science' of the day. (It was probably as apprentices to physicians and
astrologers that many Faylasufs gained access to the tradition as a whole.)
These two professions had certain characteristics in common which made
them peculiarly appropriate to the Faylasuf scholar. Both medicine and
astrology required considerable scientific training, in anatomy and astronomy
respectively; both involved in addition a set of speciously logical theories
which to us appear crudely fallacious but which appealed strongly to the
expectation that nature must form a total intelligible system; both involved
at least as much the tact and personal insight of a wise man as they did the
practice of impersonally applicable techniques.

Agrarian Age medicine based its diagnosis and prognosis on a large fund
of experience, embodied above all in the writings of the great physicians.
Of these the most honoured west of Indus was Galen, along with other men
who had written in Greek. (To the short list of classic physicians were sub-
sequently added, even in Latin Europe, such Arabic-writers as al-Razi.)
Though it was often socially impossible to do desirable dissecting on the
human body, these men maintained a tradition of careful clinical observation
which became the foundation of later medical developments.

The treatment of diseases diagnosed, on the other hand, was as always
less sure. A large pharmacopoeia of drugs was known, specifics for various
ailments. Considerable surgery was used — without anaesthetics, of course.
General rules of healthful living and of care in convalescence were urged on
the patients, with the usual results. But much medical analysis and even
some treatment was based on theories which sprang rather from abstract
reasoning than from actual experiment.

420 THE CIVILIZATION OF THE HIGH CALIPHATE

The most famous of these was the theory of the four humours, four types
of liquid (blood was the most obvious of them) which kept the body function-
ing. The four humours were held to answer to the combinations among the
four physical qualities which theory made basic to all physical things (hot
and cold, moist and dry); it was held that much disease resulted from an
imbalance in the four physical qualities in the body, which led to an imbalance
of the humours. A common application of this theory was the use of bleeding
in cases of fever (to diminish the hot blood) where we would less dangerously
use aspirin.

Most diseases cure themselves; on the other hand, sooner or later for every
mortal, however skillful the physician, some disease must prove fatal. It
was never easy to tell in practice what was and what was not effective. There
were no good grounds to reject theory that seemed to fit into the general
scheme of the natural order, so far as it could be known (and the Faylasuf
scholars hoped that human reason could know it fairly well). Accordingly,
medicine was based partly on a great deal of factual knowledge of anatomy
generally and of medical resources in particular; partly on general systematic
considerations based on suppositions about the rationality of the cosmos;
and partly on the personal skill of the physician, his sensitivity to the physical
situation and, perhaps still more, his ability to reach the patient psycho-
logically.

The case in astrology was similar. The physician was called upon to give
advice in a dangerous medical situation; the astrologer was called on to
give advice (as to timing, or even as to choice of moves, as well as in more
symbolical ways) whenever any practical situation was critical and the
outcome seemed to depend more on luck than on any calculable moves by
the individual involved. This sense of dependence on fortune emerged
especially in major enterprises such as military campaigns, but also at
marriages or at births. In each case, the ordinary person did not know how
to help himself or at least he had no way of controlling an unfathomable
future. If he took the specialist's advice and things turned out well, that was
all to the good; if things turned out ill, it was at least no worse than would
probably have happened anyway.

The ordinary person — be he peasant or king — believed equally in medicine
and in astrology. Some of the 'ulama' rejected both, as not provided for in
the example of Muhammad (except for a minimum of 'Prophetic' medicine,
mentioned in hadith, such as the eating of honey as a panacea). Most of
the 'ulama' scholars rejected astrology as diabolical (though perhaps quite
effective), but were more tolerant of m'edicine. Many of the Faylasuf scholars
made a sharper distinction, rejecting the attempt to predict outcomes by
astrology not as diabolical though true, but as actually a bogus practice.
(It was more widely admitted that heavenly movements did have effects
on human life than that humans could discern those effects.) However,
even some who personally denied the usefulness of astrology, like the great

speculation: falsafah and kalam 421

mathematician Biruni (d. c. 1050), were willing to write on it and to practice
it to satisfy their clients' demands.

Astrology was based on the doctrine that changes in the heavenly bodies
were to be correlated with changes of fortune on the earth. Yearly changes
in the course of the sun had obvious effects on human life, and changes in
the moon affected both tides and lovers. When the ancient Babylonians had
accumulated enough data about the courses of all the visible planets to
enable astronomers to begin predicting their track in the sky, including
such events as eclipses, it came to be thought that the complicated cycles
so discovered in the heavens must lie back of the otherwise inexplicable
course of earthly life — that is, of human life (for what else was so important?).
On the basis of inherited myth and perhaps some case histories, and of
reasoning by analogy, a basic set of correlations was set up; for instance,
the movements of the planet Venus — associated with the amorous goddess —
must have something to do with love. Once these basic assumptions were
accepted, sheer logic could elaborate an interpretation of every combination
of planetary positions (in particular, the planetary situation at a given
moment as it bore on the planetary situation that had presided over a given
man's birth). The required calculations were so delicate and subject to
human error that it was hard to disprove or to prove the system in practice.
It became so popular that it spread far from Babylonia as the most important
element in all the lore of divination which humans seemed to require as
insistently as they needed medicine. Along with the mathematics and
astronomy with which it was associated it became part of the Hellenic
tradition from Nile to Oxus (and in Europe), and hence a typical ingredient
of Falsafah in Islamic times.

The theory of astrology was based on assumptions about the solidarity
of cosmic events with human events, assumptions that easily recommended
themselves to people who expected the universe to present an intelligible
rational structure. Its practice, moreover, depended on that exact knowledge
of astronomy which was in any case part of the rational information which
the Faylasuf cultivated. Hence the Faylasufs could hardly avoid studying
astrology on almost the same basis as medicine. Finally, the practice of
astrological divination required the same sober temperament as did medicine.
It was demanded by clients in crises, much as was medicine, and depended
even more than medicine on the personal sensitivity of the astrologer and
on his psychological skill.

In consequence, the same man was often both physician and astrologer
and sometimes he became famous for his effectiveness in crises calling for
both types of skill. Prominent Faylasuf scholars could thus gain access to
powerful men, who protected them from the ignorant fears of people at
large (who tended to distrust them generally, and not only the alchemists,
as wizards or dealers in black arts). At the same time they were able to
gain the funds necessary to build up good libraries on scientific and philo-

422 THE CIVILIZATION OF THE HIGH CALIPHATE

sophical subjects, despite the relatively small number of those interested.
In this way, Falsafah played an important role in Islamicate society despite
the disapproval of most of the 'ulama' and the relative indifference of most
of the adibs, absorbed in genteel living and the elegant niceties of language.

Philosophy as a vision of the world

The Faylasuf, we have said, was dedicated to philosophic reason, to following
its conclusions, without yielding to preconceptions, wherever it might lead.
But 'reason' had for the Faylasufs a more exacting implication than mere
'reasoning' as a general activity. True rationality was not merely to mani-
pulate things cleverly in given particular instances (it is discussed how far
even apes can exercise such 'Yankee ingenuity'). Rationality involved
bringing all experience and all values under a logically consistent total
conception of reality. Falsafah proved to have its own special world view,
its cosmology, to which its adherents were implicitly committed. This began
with a special concept of rationality itself.

The Philosophic version of rationality required, to begin with, the acquisition
of a good deal of specialized information. But the Faylasuf scholar was not
interested just in gathering facts; he studied mathematics and astronomy,
chemistry, biology, and medicine, as rational, universal sciences. Such
knowledge, it was supposed, depended on grasping basic ruling principles —
such as the circular motion of cosmic objects, or the fourfold nature of the
sublunar elements; then all particulars, so far as they were essential to the
subject under investigation, could be logically understood. Empirical
observation (including occasional deliberate experiment) was given weight
only insofar as it promised to lead to universal principles : it was assumed
that, on the basis of a minimal (and hence practicable) amount of observa-
tion, the essential principles would emerge, from which everything could
be deduced if clear logic were held to. Even as examples of universals,
indeed, particular observations were distrusted. It was recognized (correctly,
in the conditions of the time) that in any given case too much chance entered
in for it to be a pure instance of universal rules; hence no one experience — ■
experimental or accidental — could be decisive on the level of rational
principles. The true number of teeth in the human mouth could not be
learned, for instance, just by examining an actual mouth : few people have
all their teeth, and one may always have happened on a freak. Empirical
observation, though important, could not substitute for more systematic
reasoning.

In any case, a true 'philosopher' ought not to be interested in the particular
for its own sake, as a mouth surgeon might be, studying the formation of
one man's teeth so as best to fit in an artificial one. This was of interest to
craftsmen only, skilled in an applied art, not in science fit for gentlemen.
Any concern with the time-bound, the accidental, the whole realm of the

speculation: falsafah and kalAm 423

historical, as such, was despised as unworthy, irrelevant to genuine self-
cultivation. What was wanted was an adequate understanding of the
unchanging whole; any particular instance was at best only one more
repetitive exemplification, and acquaintance with it could be of only transient
relevance, meeting needs of the moment. Some have seen a relative increase
in interest in field observation and experiment among Muslim scientists as
compared with the ancient Greeks — perhaps, with accumulating information
and technique, this became more feasible (we know too little yet to judge
well) ; but at best, the increase did not go far enough to alter the pervasive
ethos.

The model sciences of the Greeks had fitted this principle. In geometry
a whole range of propositions could be deduced from a few axioms. It was
the true triangle, which never occurs in nature, and not actual more-or-less
three-cornered objects, that could be known and was worth knowing; neglect
of the rest was what made possible geometrical calculations that were
effective even on the practical level. In astronomy, if one observed essential
regularities in a few heavenly bodies, the course of conjunctions and eclipses
could be predicted to the end of time. Ideally, all truth should be reducible
to this level of exact statement, incontestably demonstrative and timelessly
applicable (at least by approximation) to anyone anywhere. But the effort
to carry this task through required a far-reaching set of presuppositions,
which the Faylasufs had to be prepared to take as given.

A systematic admission of ignorance about crucial matters is an intellectual
luxury which non-literate peoples have rarely been able to enjoy : consequently
the best guesses available have been made and worked out to full consequences
even though no real evidence was available to support (or deny) them. Out
of this circumstance have risen the less critical conceptions of the world,
such as that which fills it with Power answering to what we feel within
ourselves, and accounting effectively for whatever may happen : if one does
not know any explanation, but dare not admit, and act deliberately upon,
ignorance, nothing makes more sense than conceptions of mana, of spells,
of exorcisms, of evil eye, of charms, of all the apparatus of animism and
magic. These are merely an extrapolation of what is most familiar. Among
the privileged classes of urban society on the agrarianate level, some were
allowed a certain intellectual leisure, that is, were in a position to devote
themselves to speculation with little thought to practical consequences.
Among them, a systematic admission of ignorance became more feasible,
and indeed essential as a prerequisite to fundamental inquiry. Falsafah
presupposed just this, and Faylasufs took pride in their emancipated views
which the masses could not afford to share.

Apart from lapses that might occur on occasion, especially in such fields
as alchemy, Faylasufs did keep free of the grosser popular errors. But there
must be a limit to one's admission of ignorance; inquiry requires not only
an unknown to explore, but some field that is known for a starting point.

424 THE CIVILIZATION OF THE HIGH CALIPHATE

For the Faylasuf, the required basic assumption was that the universe is
essentially rational — that is, rational in its essences, though perhaps more or
less unknowable in the incidental forms which these take in the coming and
going of actuality. This rationality, moreover, must be not simply an
amenability to our minds a propos of this or that particular question, but
an inherent ontological rationality, in principle independent of our minds.

But such an ontological rationality has further implications. In actual
inquiry, questions on the nature of a given thing almost necessarily include
questions on its value. If one's inquiry answered to cosmic reality at all,
it would do so on the level of valuation also. Hence throughout pre-Modern
cosmological speculation ran the expectation that different sorts of being,
different substances as such, carried a greater or lesser inherent dignity or
worth : this was not a function of one or another particular human purpose
in terms of which this or that had an intelligible value ; it was an ontological
worth independent of any human consideration. A plant was higher in the
scale than a stone, an animal than a plant, a man than other animals, an
angel than a man. Such a conception seemed to be a natural consequence
of the ontological rationality of the universe as a whole: if the rational
interconnections of events are to be ontologically inherent in the structure
of existence, then the values of given events, without which we cannot
really think of them or reason about them, must be ontologically inherent
likewise.

If, then, this ontological rationality and value-structuring of the universe
be accepted (as that essential starting point for inquiry, beyond which a
systematic admission of ignorance becomes self-defeating), a total world view
can be built up in which everything has its place. Falsafah, as inherited
from the Greek schools, had developed this viewpoint very consequentially.
All knowledge, from the most special sciences to the most general synthesis,
formed a single whole ; in particular, theories in any one field derived their
premises from the conclusions of other fields. Just as theories in medicine
depended on the results of investigations in physics, so theories in physics,
in turn, depended on metaphysical conclusions, as did the principles of
ethics or of aesthetics. To attack the edifice at any one point was to attack
the whole.

Being was composed of entities which might act on each other. Each
entity was made of fixed substances, characterized by changeable accidents —
as a piece of wood might be painted green or blue. What one sought to
explain was not events or situations, but the distinct entities as such: the
word answering to our word 'cause' meant not a cause of an event, but a
constitutive principle of an entity — the material it was made of, the form
it took, the purpose it served, and the agent (itself an entity) that occasioned
it. (Hence, in contrast to systems presupposing flux or process in a continuum
of fields and gestalts, movement and change were seen as something super-
imposed from without.) Each type of entity had its tunelessly essential

speculation: falsafah and kalam 425

character, without which it would be something else, and which could be
formulated in a categorical definition: wood was in the generic class of
earthy solids, singled out specifically from other such solids by its fibrous
character as the densest part of a growing plant. The human being was
an animal, specifically a rational one. And to each species, each type of
entity, could be attached an equally changeless inherent worth or dignity.

The entities, as studied by physics, were mostly made up of compound
substances — compounded of the four elements, with characteristics derivable
from the nature of their compounding (chemists attempted to find what
reactions could produce what compounds). The purest compounds — that is,
those coming closest to an ideally harmonious combination of the elements —
were endowed with greatest dignity. The four elements, ideally, were located
in concentric spheres, earth at the centre, then water above it, then air,
and fire highest of all: their mixture was already impurity and occasioned
all the motion they were constantly in, trying to regain their natural position.
Astronomers claimed, however (though not all did so), that it was only the
substances in and on the earth, in the sphere below the moon, that were
compounds, and hence perishable, subject to coming into being and dis-
integrating, to generation and corruption. From the moon's sphere upward,
the changelessly circular nature of the heavenly motions showed that they
were of 'ethereal' stuff purer even than the four elements here below — which
latter moved only when they had been displaced, and then linearly, and
hence not indefinitely.

The Faylasufs' notion of God

Given this conception of the material universe, it was natural to proceed
to a systematic account of the cosmos as a whole. In Aristotle's system,
metaphysics was merely what one could study once one had mastered the
principles of physics; it was basically the same sort of study, with the same
sort of consequences. It is especially clear in such inquiry, how far this sort
of rationalistic analysis could go in discussing questions on the most complex
human levels; and what its human implications were.

These presuppositions about being underlie the proof of the existence of
God given (on the basis of earlier writers) by Ibn-Sina. (Avicenna — d. 1037),
the most famous of the Faylasufs. For those to whom 'God' meant, in the
first instance, not an experienced challenge but a cosmic entity, as such
not directly experienced, its very existence had to be demonstrated. Ibn-
Sina's proof assumed two features of the Philosophic world view. The world
is made up of composite things, in which our minds must distinguish more
than one component — not only compound entities, but even the simplest
are composite, distinguishable at least into form and matter. Yet our
reasoning looks ultimately to something not composite but truly simple —
something that shall be irreducibly itself. We speak of a tree, a stone, water,

426 THE CIVILIZATION OF THE HIGH CALIPHATE

air; but if we find these things can in turn be analyzed into component
parts (bark, wood, pith; silicon, mica, quartz), we know we cannot rationally
understand them till we take into account their components, reducing them
to ever simpler elements till we can no longer subanalyze them. If, then,
such rationality is conceived as not merely a function of the way our minds
work, but as defining the ontological nature of existence, we will find that
the simples will seem primary and the compounds derivative; likewise,
what is at rest will seem primary and what is changing will seem derivative.

In this case, it makes sense to say that anything composite must not
merely be accounted for in terms of our minds, but must be admitted to be
caused by something, ontologically, and therefore to be contingent upon
something outside itself. Whatever is contingent has a lower status, a lower
dignity, than whatever is self-dependent, just as in a family the children
depend upon the father's wishes and are ordered about by him and so have
a lower dignity than the father, who does as he wills and is not controlled
by, or dependent upon, the children. Only what is absolutely simple, then,
will not be caused by something else and hence lower in dignity than it;
what is absolutely simple will be a necessary being — that is, it comes into
being through itself. Is there then such a 'necessary' being, in the sense
of something simple and uncaused? It is tempting just to say that since
such a being is 'necessary', it needs must be. But in this context what matters
is ultimately that a rational hierarchy of entities requires one ultimate
entity at the top of the hierarchy; a sequence of caused beings requires an
uncaused being at the end to start the causes, lest we find ourselves in a
rationally intolerable infinite regress.

In what sense is such a first cause of all being to be identified with what
in the monotheist tradition was called God? A metaphysical first cause is
not necessarily the same as a Creator-god. Yet the Faylasufs' 'necessary
being' will have the highest possible ontological attributes — be the most
worthy and excellent of all beings; for it will be perfect (being self-sufficient,
not dependent upon anything else). And if it is perfect it will not be lacking
in any valuable quality; not only valuable in that we regard it so, but
inherently valuable in the rationally ordered universe: hence it will know,
if anything knows ; it will be alive, if anything is alive ; it will be most powerful
if anything has any power — it will, then, be a most excellent and perfectly
admirable being, worthy of worship and honour by any other, more than a
strong king is honoured by his subjects or a good father by his children. The
Faylasuf could feel that such a being was worthy — and alone worthy — of
the status the monotheists ascribed to God.

In later Roman times, a group of Greek Philosophers (called neo-Platonists) ,
led by Plotinus (d. 270), had developed some of Plato's insights into an
elaborate cosmology relating this First-Cause God to human life in a way
that made religious sense. The neo-Platonist cosmology rested on three
bases : on Aristotle's astronomy and metaphysics, which pointed to a cosmic

speculation: falsafah and kalam 427

first cause in rather the way we have seen above; on some of the logical
and mythological conceptions of Plato, who had been interested in the
aspirations of material beings toward non-material norms beyond them;
and on mystical experience. The neo-Platonist cosmology originally pre-
supposed the pagan Hellenic religious notions, but could be adapted to
Christian doctrines; it was generally accepted in the tradition of Philosophia
by Islamic times. (We have already seen traces of it in the Isma'ili system.)

As the Muslim Faylasufs developed this cosmology, it was a timeless
drama of the unfolding at once of logical and of material potentialities. The
universe originates eternally, timelessly, from an absolutely transcendent
One, the absolutely Simple, of which nothing positive can be properly
predicated except what is necessary to define it as the First Cause. Just by
the rational purity of its oneness, however, emanates from it eternally
Absolute Reason, rationality as such — still a single thing, but carrying
within it implicitly the multiplicity of harmonies and relations. From the
confrontation of the Absolute Reason with the Absolute One can arise,
then, all that goes with multiplicity — movement and change; and in parti-
cular, the Absolute Soul — the ultimate principle of self-motion. Soul produces
motion (and so time) in its necessary response to Absolute Reason, its
attempt to imitate the perfection of Reason; which it cannot achieve,
precisely because its nature is derivative from Reason and not reason itself.
From Absolute Reason and Absolute Soul proceed the concentric heavenly
spheres of the Aristotelian and Ptolemaic astronomical systems, and their
pure circular movements ; each of these spheres, from that of the fixed stars
through the several planetary spheres to that of the moon, has its own
Reason and its own Soul, governing its motion. The interaction of all these
circular movements, which become more complex as they descend toward
us, accounts for the complexity of the sub-lunar sphere, the earth with its
four elements and the diverse compounds thereof. (Over all this complexity
presides the Active Reason of the lowest sphere.)

To this point, the theory merely accounts for the diversity and movement
we find in the observed universe of entities; but its religious meaning emerges
when it is seen that human reason is a reflection of that Active Reason that
presides over our world. If we purify our own rationality, we can participate
positively in the whole process, thus reversing in our own consciousnesses
the descent from unity to multiplicity by reascending intellectually from
multiplicity to ultimate unity. In this way, the whole Philosophic life is
justified: the search for intellectual awareness, and in particular for under-
standing of the universal as against the mere transient instance; and the
moral purification of the self which at once makes dispassionate rational
inquiry possible, and is its outcome. The Philosophic search is the truest
way of honouring and worshipping God; the cults and moral rules and
doctrines of ordinary ignorant people are merely imperfect attempts at the
true Philosophic way.

428 THE CIVILIZATION OF THE HIGH CALIPHATE

Such a religious vision could be harmonized, to its own satisfaction, with
almost any popular religious cult. But the exponents of alternative visions
might not be satisfied with the terms of the harmonization. The monotheists
of the Irano-Semitic tradition, in particular, could doubt if the Philosophic
One First Cause were God at all. The monotheists conceived their God as
having rather different fundamental attributes, both as supreme Creator,
creating by an act of arbitrary will, and as final moral Judge, intervening
positively and personally in human life. The monotheists' notions of God
had been built up precisely from observing and responding to those con-
tingent and historical data which the Faylasufs tended to disregard as not
amenable to reason. The prophets' idea of God was more moral than onto-
logical, more historical than timeless; God might want honour, but first of
all He wanted obedience, and the honour would follow if He chose to exact
it. It had long been evident that the Philosophic and the Prophetic traditions
were not in ready harmony. In Islamic circumstances, as earlier, it was
not easy for them to come to terms with each other. 5

The Faylasufs' 'God' remained a very different figure from the God of
the prophets, as different as their sense of human destiny; and however
much the difference was disguised by the use of common words, it showed
up at crucial junctures. The Faylasufs, above all, rejected any cosmic history
with a beginning and end; hence they could not accept God as Creator, as
Providence, or as Judge, in the Abrahamic sense. Instead, for them, the
world was eternal; God (as ultimate Reason) could take cognizance only of
abstract 'universal' intelligibles; and He certainly would not arbitrarily
resurrect human bodies at some end of time. It was at such inescapable
points that it became clear whether a person's first allegiance was to the
Abrahamic or to the Philosophic tradition.

Falsafah and the Islamic revelation

The adepts of Philosophia saw it as a necessary consequence of pure reason,
the same for any rational being of whatever nation or time. (Philosophia
was felt to be simply philosophy, in the sense of true understanding and
wisdom.) When, like all other cultural elements, Philosophia too was trans-
lated from older languages into Arabic, nothing new need have been anti-
cipated from it. But, as. we have already seen in the natural sciences,

5 An excellent presentation of the assumptions and methods of basic inquiry and
its pervasive moral implications, as the Faylasufs saw it, is al-Farabi's Philosophy of
Plato and Aristotle, translated and introduced by Muhsin Mahdi (Glencoe, 111., 1962).
Note that though, as Mahdi points out, the neo-Platonist system of emanations from
the One is not mentioned even in the first portion (which shows what any proper
philosophy must do — the latter portions exemplify this in the two master cases — and
presents very few other positive doctrines either), yet the emanations are clearly pre-
supposed in more than one place, and the basis in method that leads to them is clearly
stated, especially where the study of stars and of souls is made the double point of
departure for studying the divine principles in metaphysics.

speculation: falsafah and kalam 429

Islamicate Falsafah proved to be more than a mere continuation in Arabic of
the Syriac tradition. The burgeoning society of the 'Abbasi cities gave a
strong impulse to rational inquiry; but the new intellectual constellation of
Islamicate society also had an effect. The results were historically distinctive
even in what we would call philosophy proper.

Despite its rejection of the historical and the traditional, despite its
attempt to deduce everything from abstract and timeless rational principles
Falsafah itself was, of course, a historical, cultural tradition. It had had its
own charismatic founders — most notably Socrates, to whose memory and
example Faylasufs were loyal. And while Faylasufs made a point of their
independence toward earlier authorities, quoting Aristotle's point that he
owed great friendship to Plato but greater yet to Truth, they inevitably
studied such masters and had to come to terms with their thinking, even
when they tried to improve on it. Under the new circumstances introduced
by Islam, new questions and new opportunities arose and the Philosophic
dialogue was quickened: Falsafa, too, was developed through the interplay
of conscious individual decisions when the social contact raised new problems ;
and the outcome of those decisions affected the whole subsequent tradition
of Falsafah, setting limits to what Faylasufs could attempt to do for the
future in relation to society, and hence to the kind of viewpoints they could
relevantly develop. 6

As we have seen, the Philosophers of the Greek tradition had gradually
developed a certain degree of unity in their general approach, taking the
logic and a good deal of the particular sciences from Aristotle, while the
picture of the cosmos and the place of the individual in it went back largely
to Plato or to Plotinus' interpretation of Plato's thought. Within this general
orientation there was considerable variety in the schools in the emphasis put
on Aristotle or on Plato; and indeed other strands of thought, notably the
later Stoic, continued to command attention. But each student was expected
to read a certain set of works with his masters and in almost a set order.

It was in this more or less standardized form that the Philosophic tradition
appeared at the time when its most important works were translated into
Arabic, and it was from this point of departure that the Muslim Faylasufs
began to speculate. However, when the classics of philosophy and science

6 The contrast sketched in this chapter between the Prophetic and the Philosophic
traditions has been standard in the literature for decades now. It has played somewhat
comparable roles in Islamdom and Christendom. Often it is identified with a contrast
of Semitic, or at least Hebraic, culture vs. Greek. The contrast is so neat that it is
always in danger of being pushed too far. This danger is at the worst when it is made
to look like a contrast in 'racial' temperaments. In interpreting it as a contrast between
historical dialogues, each of which is (by the nature of dialogues) open to interinvolve-
ment with the other, I have hoped to minimize the danger. In particular, it becomes
clear that each dialogue has been an integral part of the total culture both in the Irano-
Semitic complex and in the European, even though each dialogue has played a some-
what different role and even carried somewhat different elements in the two regional
cultures.

430 THE CIVILIZATION OF THE HIGH CALIPHATE

were translated into Arabic, this tended to be done piecemeal for the benefit
of individual laymen and scholars who could not trouble to learn Greek. In
the process, the old system of a set curriculum was partially interrupted.
Those who learned their Falsafah only or primarily in the Arabic translation
were likely to come at it arbitrarily ad libitum. This in itself was enough to
suggest new perspectives: when the problems were not reached in the
standard order and explained by the standard commentators, new facets of
them became obvious, and the Islamicate Faylasufs were led to present the
material in new, independent syntheses.

Thus Muhammad b. Zakariyya' al-Razi (d. 925 or 934), the physician, was
moved, evidently having read a number of earlier philosophers not normally
included in the Philosophic curriculum, and having become intensely aware
of the gaps in intelligibility still left by the standard views, to reject a cosmic
system in which there was just one eternal First Cause, 'unmoved mover'
eternally setting in motion the otherwise unformed matter of the universe.
Instead, he postulated five eternal first principles, co-existent and possessed
of equal logical necessity. He could not see how time or space or matter
could either be generated from another intelligible first principle (Reason or
Soul), or be subsumed under any other category. His picture of the cosmos
presupposed all the basic assumptions that Faylasiifs had to take for granted,
and in this sense marked a relatively slight new departure in the tradition,
which was as powerful in him who so severely criticized the old metaphys-
icians as in those who departed less noticeably from their solutions. But
under the new circumstances brought by Islam (though he was at most but
a nominal Muslim), al-Razi was led to deduce from the general framework
some quite novel conclusions.

But the most important consequence of the Islamic environment was the
new confrontation with the monotheistic tradition which it enforced.
Though some Faylasufs clearly accepted no regular religion whatsoever,
most Arabic-writing Faylasufs were either Christians, Jews, or Muslims;
they all acknowledged the pagan Greek sages, especially Plato and Aristotle,
as their masters, but their doctrines on crucial points did vary according to
the allegiance they held. Christians had worked out their solutions to the
problem of reconciling Christianity and the Greek intellectual disciplines
some time since. But Islam presented a more concentrated challenge. The
Muslim solution was only in part parallel.

The first of the philosophically independent Faylasufs of Muslim faith
was Ya'qub al-Kindi (d. after 870 — by way of exception among Muslim
scholars, he was of old- Arabian descent). He adapted to the Islamic doctrine
with little change in spirit the Christian solution which his teachers had
accepted, that revealed teachings about God and the soul were parables of
Philosophic truths. He seems to have worked closely with Mu'tazili thinkers
in their attempt to eliminate anthropomorphisms from their faith, and to
have believed that their solutions were not incompatible with rationalistic

speculation: falsafah and kalam 431

philosophical views. He was especially known for his collection of philosophical
definitions and was perhaps more concerned with encouraging a general
rationalistic viewpoint than with working out any particular system. But
this approach could not remain satisfactory. The Socratic tradition could not
rest content with being bound to limit its questioning within a framework
which was imposed by a historical intervention such as Islam. Nor could the
Qur'anic tradition accept subordination fo its conclusions to the authority
of private human speculation.

The alternatives of Socratic faith and Abrahamic faith

The dilemma is illustrated in conversations recorded by Abu-Hatim al-
Razi, an Isma'ili da'i preacher, as taking place publicly between himself
and the Faylasuf Razi, Ibn-Zakariyya.'. 7 Of all those who shared the Qur'anic
tradition, the Isma'ilis were the most actively interested in the doctrines
associated with Falsafah; their symbolism presupposed the old Hellenistic
image of the world, and in the following centuries they were to adapt the
old Philosophic cosmology to their own purposes. But the Isma'ili was as
shocked as any other adherent of a monotheistic revelation at what he
regarded as the dangerous and arrogant rejection of historical revelation
recognized by the community, in favour of the momentary speculation of
an individual.

The Faylasuf Razi began by objecting to reliance on revelation on the
ground that, since more than one doctrine is supported by revelation and
there is no way to decide between the conflicting claims, such an appeal can
only lead to destructive conflict among mankind — which cannot be God's
purpose. All humans alike being endowed with reason, this alone has any
hope of settling disputes, and it must be God's intention that we use it for
reaching truth. The Isma'ili Razi retorted that while we all have reason, we
do not have it all alike; some persons are so much more intelligent than
others that it is inevitable that some lead and that others follow. Hence
there must be an ultimately authoritative leader if truth is to be found at
all — as surely God desires it to be; this leader (to be identifiable by the less
intelligent) must be a prophet bringing revelation.

But then he turned the Faylasuf's point against him, noting that in fact
the proponents of individual reason differed among themselves just as did
the proponents of revelation. He noted that the Faylasuf Razi had himself
condemned important conclusions of great earlier rationalistic Philosophers
and asked if he, their disciple, thought himself wiser than they who had
taught him — implying that even the wisest Philosophers make mistakes and

7 Both were from Rayy, hence called Razi. I summarize here from 'Munazarat bayn
al-Raziyyayn' by Abu-Hatim al-Razi in Muhammad b. Zakariyya.' al-Razl, Opera
Philosophica, ed. Paul Kraus, Fu'ad University Faculty of Letters, Publication no. 22
(Cairo, 1939), vol. 1, pp. 291-316.

432 THE CIVILIZATION OF THE HIGH CALIPHATE

have no way of coming to a final settlement. The Faylasuf declared that the
later thinker, having the benefit of all that the earlier ones had done, could
add his own inquiries and so improve on their work; on which the Isma'ili
pointed out that there would always be still later thinkers to improve further
on al-Razi's own improvements, so that al-Razi had no assurance that he
was right; and (since in fact the various opinions were all retained in the
books side by side) the result was simply a multiplication of the number of
diverse opinions about which people could dispute.

But at this point the Faylasuf turned to the heart of the matter. He was
not immediately concerned with an indefinite progress in knowledge for the
benefit of hypothetical future generations (when supposedly the range of
diverse opinions might have been narrowed by exhaustive selection). Rather
he was concerned with the pursuit of truth in any one generation. He main-
tained, then, that what counts is that each thinker be putting forth his own
best effort; he will then be on the way to truth, even if he does not attain it,
and it is being on the way that is desirable. For what is wanted is to purify
the soul of its 'turbidness', of the confusion of mind induced by the sensory
impressions and passions of living, so that it can judge and act objectively.
And 'souls are not purified of the turbidness of this world or freed for that
other world [the life of spirituality]' except through independent study and
examination ; 'when someone studies it [rationalistic Philosophy] and attains
something of it, however little it be, he purifies his soul from this turbidness
and renders it free. And if the common crowd who ruin their souls and are
neglectful of study would devote but the slightest concern to it, this would
liberate them from this turbidness . . .'

The Isma'ili, however, convinced of human intellectual inequality, was
less optimistic about the common crowd. He asked if a person who continued
to believe in the doctrines of revelation could be purified by studying
(rationalistic) Philosophy a little on the side; to which the Faylasuf
responded that no one who persevered in submission to established opinion
could become even a student of 'Philosophy' (thus indicating what he really
was demanding when he asked for even a little bit of independent study).
But the Isma'ili pointed out that, in practice, those who went into ration-
alistic Philosophy less than very deeply (which would be most of the would-
be rationalists) might indeed reject revelation, but only to submit to a
different tradition — a different set of established opinions — those to be
found in the books of the Philosophers, which the Faylasuf had admitted
were not necessarily in themselves true. Such a person would lose the benefit
of the historical revelation without achieving the purification the Faylasuf
called for; none could be in a worse state than that.

Both the Faylasuf and the Isma'ili were concerned with the moral dimen-
sions of living, to which sheer knowledge was only an instrument. The
Faylasuf Razi looked to the process of inquiry to make a good man, and was
(exceptionally among the Faylasufs) ready to see a relatively large public

speculation: falsafah and kalam 433

join in this process — he wrote a book on treating the ailments of the soul,
aimed partly at persuading the adib to look beyond his polite and superficial
culture. The Isma'ili Razi demanded, as a good Shari'ah-conscious Muslim,
that responsible living be based on something more objective than on an
inner 'purity'; he looked for the most tangible possible assurance of a
socially valid position — which meant turning to a historically accepted
revelation, that of Muhammad and the Qur'an. Most Faylasufs acknowledged
that, as regards the common masses at least, the Isma'ili had a point.

Al-Fdrdbi: revelation as political truth

To the problem, how the rational potentialities of human beings are to be
fulfilled without leading them into mazes of subjective and anti-social error,
most Faylasufs had long since found a solution on the practical level. To
begin with, they minimized the differences in opinion among Philosophers.
We have noted how, already in Greek and in Syriac, the standard Philosophic
curriculum was based on both Plato and Aristotle, in different respects,
despite the well-known contradictions between their views. Unlike al-Razi,
most Faylasufs continued relatively close to the old syntheses. Though the
differences among the old masters were not ignored, Philosophia was felt to
offer a single orientation and purpose, and hence, at least potentially, a single
doctrine. Explicit harmonization between the two greatest masters was
furthered in two ways. Plato's logical method, for instance, was interpreted
by way of Aristotle. This simply reflected a straightforward attempt to solve
the intellectual problems both had left open. But some harmonization was
in direct response to the need to justify Philosophia as not contradicting
revealed religious dogma. Thus Aristotle's teachings that seemed inconsistent
with immortality for the soul and with a temporal act of creation were
interpreted by way of Plato, sometimes with the aid of a neo-Platonist
booklet early translated into Arabic and mistitled 'the theology of Aristotle'.
But this assurance that true Philosophers did come to the same conclusions
in their independent inquiries was only part of the answer to the danger of
vulgar confusion resulting from too free a use of individual reason. The
Faylasufs' more basic solution was to preserve rationalistic Philosophia for
themselves alone, and to sacrifice any possible human universality of
rationalism in cautious recognition of the inadequacies of ordinary people.
This meant coming to terms with popular revealed religion.

The man who established the classical tradition of the Muslim Faylasufs'
attitude to revelation was Abu-Nasr al-Farabi (d. 950). Al-Farabi, born on
the Syr river in a Turkic military family, studied at Baghdad and lived there
on private means until late in life he became court musician at Aleppo.
Unlike Ibn-Zakariyya al-Razi, al-Farabi took Islam seriously. But he had
studied under a school more uncompromisingly rationalistic, among the
Hellenic Christian traditions, than did al-Kindi, as can be seen from the

o
o

o

Xi
ft
o

Oh

I

° 2 _■

6 g °

2 o

.a ta

b "u
2 >

42 s
fc —

s

o 5?

S 2
&
ui fn

« 2
v G
Xi O

+> rB

w -a
11

xi
+j

-d

+i

->£
rt

a;

,B
w

B

■c

o

V j3

■a « S

'53 o m

ni tfi J3

"" .2 y,

a- a.

nj

ctS

■5 >

0)

+T rr-t

B K

B o

Is

2 S.
" §

■u

a

73

p<

B

3-3

.a ^

<i .2

B
OS

s «
►2.2

XJ aj
-> m h-

,B 2

$?£

« B

° .2
73 ^-,

O l- 1

. „

s

42

"«

Y-i

H-l

en

o

"^p

•i — ■

to

<u

a

h

l-l
o

X>"

B

in

p

N

a*

60

X)

JA

cfl

o

1-1

in

ni

9

<a

,Q

■s-

-B

,2
3

o
73

#

ni

H-l

S

v*

«-l

+3

_o

B

X>

S

M

o

o

+j

+j

o

o

B

B

O

W
O

xi

,B

o

*^

ctj

-H

o

>,

>>

B

CT>

CO

M

£

u

-B <*

xi S
B <d

rt <!3
?0

H O

S x)
2 -«
S o
5 o

&^

^

+J (U

g>

.5 £

t S

o
to

o

C« i^ 1

***

B 73

C^

^ .2

M h0

■&>

■ - o

•I" 3

^

CO i-t

O

B ft

speculation: falsafah and kalam 435

chart of early Islamicate Philosophic schools. Then he added a powerful
perceptivity of his own. 8

Al-Farabi and his followers seem to have had a more developed concern
with social or political questions — especially those where religion entered
in — than Christian (or most ancient Pagan) Philosophers had had; the
political orientation of Islam itself suggested this, but the general cultural
fluidity of the times perhaps suggested, even apart from Islam itself, the
revival of Plato's political hopes which he had expressed in Sicily and which
his disciples had expressed somewhat more successfully in new Greek
foundations. Al-Farabi, and no doubt others, perhaps felt that with the
coming of Islam and its universally accepted godly ideals a new situation
had appeared, which might be favourable to building a philosophically more
ideal society. At any rate, al-Farabi and his successors spoke of gradual
reform and put some of their best efforts into political philosophy.

The phenomenon of Islam was too massive, whatever a man's personal
feelings toward it, to be ignored by one who would understand the universe
and human life in it. Why did so many human beings adopt this interpre-
tation of life and society? What was its meaning and its worth as a system
of thought and action? Already Plato had suggested that a well-ordered
society required doctrines which would be supposed by the population to
be divinely revealed, doctrines which were not necessarily true in themselves
but would lead the people to behave in the ways that were best for society
and for themselves. Al-Farabi adapted this point of view to Islam. 9

The inhabitants of a good society ought to believe that there is a single
God — though they need not be expected to have a full rational— i.e.,
Philosophic — insight into that God's nature ; and to believe that humans, as
beings endowed with reason who should rise above material preoccupations,
should, in respect before God, behave justly toward one another — though,
again, they need not understand rationally the nature of either human beings
or of justice. These basic beliefs could be put before the masses in the form
of images, such as the depiction of Hell as a suggestion of the subhuman
fate of the person preoccupied with material lusts. Ideally, perhaps, a good
society should be ruled by a perfect 'philosopher', a sage in the Faylasufs'
sense, who would understand all its principles himself as well as know how
to inculcate the needful ideas in others. But if this were impossible, various
approximations would do; at least the laws should have been constructed

8 Useful general articles on al-Razi and al-Farabi, by Abdurrahman Badawi and
Ibrahim Madkour, respectively, appear in M. M. Sharif, ed., A History of Muslim
Philosophy (Wiesbaden, 1963) — a very uneven collection of disparate pieces.

' Leo Strauss, 'How Farabi Read Plato's Laws', Melanges Louis Massignon,
(Damascus, 1957), vol. 3, pp. 319-44, gives an instructive example of Strauss' way of
reading the Faylasufs to bring out this viewpoint; the method can seem unconvincing in
one sample, but cumulatively it comes to impose itself. The alternative is to suppose that
al-Farabi and all his admirers were fools. It must be added that Erwin Rosenthal, in
his various writings, displays a consistent incomprehension of the Faylasufs' political
thinking.

436 THE CIVILIZATION OF THE HIGH CALIPHATE

in accordance with rational, philosophic requirements and presented in a
form which would induce the unphilosophical people to observe them. The
role of Muhammad, then, was that of lawgiver: his great mission was to
bring the Shari'ah law and to surround it with sanctions which would cause
it to be observed.

Popular religion, in this view, was more than an expedient fiction, though
it was also not wisdom to be placed on a level with rational 'philosophy'. 10
Revelation was a natural process essential to the constitution of society,
which could not exist without some form of imaginatively sanctioned law.
The Prophet was bringing truth, in the form that he had the gift of bringing
it in, even though it was not so high a truth as was open to the rational
'philosopher'. Its highest doctrines were statements of basic rational
principles, but put in a simple or an imaged form which ordinary people
could understand. The law which the Prophet promulgated was valid and
genuinely incumbent on those human beings who happened to be within the
community for whom he was legislating.

There might be little reason, in the abstract, for a rational 'philosopher'
to prefer one popular religion over another. But the Faylasuf might feel that
Shar'i Islam, with its emphasis precisely on law and on the social body and
its relative freedom from such excessively fanciful dogma as that of the
Trinity, met the rational requirements for a good popular religion with
singular purity. Al-Farabi used language that suggested (at the Shi'i court of
Aleppo), by stressing the importance of an imam sound by nature as the
continuing head of the community, that not just Islam but Shi'i Islam
might be looked to to bring about a more satisfactory society. In any case,
the rational 'philosopher' was philosophically bound to adhere to the popular
religion of his own community, acknowledge its doctrines (however he might
tacitly understand them), and support its practices. The religious learning
formed an integral part of what a 'philosopher' ought to know — under the
heading 'political philosophy'. In this limited sense, a Faylasuf in Islamdom
not only could but should be a good and believing Muslim.

Yet the traditions of Falsafah were not only older than and independent
of Islam ; they presupposed a radical elitism, in which only those could share
who disposed of the material as well as the intellectual leisure to devote
themselves to pursuits that had little practical application. Even such
utility as the Faylasufs could in fact offer was restricted to the wealthy,

10 Many scholars have supposed that al-Farabi was trying to reconcile 'religion and
philosophy', 'faith and reason', as two independent sources of the same truth. He did
want to show that nothing which the monotheistic tradition insisted on had to be
rejected by rationalistic philosophy; but unlike al-Kindi, who might indeed have
conceived the question in some such terms, al-Fa.ra.bi and the later Faylasufs restricted
the spiritual role of popular religion very sharply and gave it very little intellectual
credit in any case. R. Walzer, Leo Strauss, and Muhsin Mahdi have been the most
helpful scholars in pinpointing the Faylasufs' attitude. I have learned a good deal from
Mahdi orally, though I have not been willing to accept some of his viewpoints.

speculation: falsafah and kalam 437

generally monarchs and courtiers, who could afford their expensive minis-
trations. Shari'ah-minded Muslims always felt the Faylasufs as alien. Their
reaction was taken up by the man in the street, and Falsafah as a whole,
together with the sciences which it specially fostered, was never fully
accepted in High Caliphal society as an integral part of culture, in the way
that both the prophetically-based Shari'ah and even the absolutism were
accepted. It remained a bit apart and foreign.

Earlier Translators, Philosophers, Scientists

fl. 770s Jabir ibn Hayyan, alchemist, eponym for Jabir corpus of

alchemical knowledge
c. 800 Teaching hospital established in Baghdad

813-848 Mu'tazilism flourishes, intellectual endeavor of all kinds heavily

patronized by the court
c. 850 Death of al-Khwarazmi, 'father of algebra', mathematician

(user of the 'Indian' numerals) and astronomer
c. 873 Death of al-Kindi, often called 'the first Faylasuf; death of.

Hunayn ibn Ishaq, important translator of Hellenistic works

into Arabic
Early tenth 'Jabir corpus' of alchemical knowledge in circulation
century

925 Death of al-Razi (Rhazes), Faylasuf, alchemist, physician

929 Death of al-Battani, mathematician (trigonometry), astronomer

950 Death of al-Farabi, Faylasuf-metaphysician

c. 970 Collection of 'Epistles' (Rasa'il) of the Ikhwan al-Safa', a

comprehensive compilation of esoteric (Pythagorean-type)

'scientific' and metaphysical knowledge

Kaldm and Falsafah

The glory of the Faylasufs was their cosmology, their rationalistic portrayal
of the universe as a whole and of the place of the human soul in it. This
was a standing challenge to the devotees of the monotheistic tradition to
do as well. The Piety-minded had tended to develop, as part of their over-all
intellectual structure, the rudiments of an argued theory of God and mankind.
At first, this was done point by point at need; for instance, where the Islamic
revelation was challenged by other monotheists: the absolute unity of God,
the prophethood of Muhammad, and generally the validity of Qur'anic
descriptions of God and of the Judgment. The first aim of the Mu'tazilis,
who first cultivated extensively this sort of apologia, was negative, to show
that there was nothing in the Qur'an which was repugnant to careful
reasoning. But very soon, in the course of disputes with non-Muslims as
well as with Muslims (such as the subtle anthropomorphist — and Shi'l —
thinker Hisham b. al-Hakam), they had to decide standards of what should
be considered repugnant to reason; this ultimately meant establishing an

438 THE CIVILIZATION OF THE HIGH CALIPHATE

overall cosmology which they could claim was rational and with which
they could show the Qur'an to be in harmony.

Such activity was called kaldm, 'discussion', that is, discussion of points
of religious belief on the basis of rational criteria. The more exacting of the
Shari'ah-minded regarded this as bid'ah and denied that human minds
should presume to 'prove' what was given by revelation, but the activity
became widespread as Islam became the prevailing religious allegiance, and
it gave rise to several competing theological schools.

The first great systematic thinker among the Mu'tazilis, who gave that
school of piety, already devoted to argumentation, the framework of a
comprehensive dogmatic system was Abii-1-Hudhayl (d. 840). Among his
followers, doctrines were worked out under five heads: (1) the unity of
God, under which heading Manichean dualism, but also anything that
smacked of anthropomorphism, was condemned — and in particular the
cruder literalism with which the Hadith folk took Qur'anic images ; (2) the
justice of God, under which any ascriptions of injustice to God in His judg-
ment of human beings was rejected — with the consequence that human
beings were made alone responsible for all their acts, and so punishable for
their bad ones; (3) the coming Judgment, under which the importance of
righteousness was insisted on, as against mere Muslim allegiance — with a
tendency to exalt the Qur'an as the standard at the expense of hadith; (4)
the intermediate position of the Muslim sinner, neither faithful nor infidel;
finally, (5) the obligation to command the good and forbid evil — to take
initiatives toward godly social order. Within this framework, a wide range of
differing opinions was to be found within the school as it developed.

It was under the heading of the unity of God that the Mu'tazilis raised the
most cosmological questions, for they had to find an understanding of God's
creation and governance of the world: was there, for instance, anything
that God could not do? — which most denied. Their terminology — substance,
accident, existence, non-existence, etc. — seems to echo, though imprecisely,
the Greek philosophical tradition; perhaps it reflects Christian theological
discussions with which the Mu'tazilis and other Muslims took issue. But
their orientation was to the primacy of revelation.

Mu'tazilism was never the only school of Islamic speculative thought, but
for long it was the most intellectually effective. However, with the rise of
the piety of the Hadith folk, it found itself increasingly out of step with the
most popular currents of the time. The metaphysical unity and the rational
justice of God could be defended out of the Qur'an and were consistent
enough with the relatively imprecise 'faith of the ordinary Muslims of early
times. But on these as on almost all their main points the Mu'tazilis were
at odds with the Hadithi movement, which tolerated seeming anthropo-
morphism, stressed the omnipotence of God more intensely than His justice,
and accepted hadith reports often at the expense of the Qur'an. By the end
of the ninth century several men, including some who identified themselves

speculation: falsafah and kalAm 439

with the Hadith folk, had begun making efforts to justify rationally, against
the Mu'tazilis, these more popular views — departing from the more common
position of the Hadith folk, of rejecting all kalam disputation altogether.
Two of the most prominent were Abu-1-Hasan al-Ash'ari (d. 935) and
Abu-Mansur al-Maturidi (d. 944).

Such men attempted to limit kalam and its cosmology more closely to a
defence of the doctrines given in the accumulating hadith, which now
represented the more generally approved doctrines of the Shari'ah-minded.
But their concern was not simply with individual doctrines, but also with
the temper of the whole intellectual movement. Sceptical of the powers of
abstract reason, they tried to deduce as little as possible from the supposed
requirements of reason as such, as the Mu'tazilis had often done. To this
extent, they backed away still further from the viewpoint of the Hellenic
tradition even while retaining its categories. 11

What positions were actually adopted by al-Ash'ari is not yet clear; as
eponym of a school of thought, he has been ascribed all the basic positions
taken later by the school ; and not all the works bearing his name are genuine.
But it was probably al-Ash'ari himself who developed the key formulations
on some of the issues most mooted by the Mu'tazilis. The Mu'tazilis had
insisted that to describe God as possessed of any distinct attributes at all,
such as knowledge or power, was to run the danger, if not of anthropo-
morphism, then, worse yet, of asserting other eternal entities with God: for
any knowledge or power attributed to him must be, as such, eternal. They
insisted that God knows or is powerful not by a special knowledge or power
but by His simple essence in each case. Doubtless unconsciously, they were
here allowing the very notions of 'attributes' and 'essence', inherited from
the Philosophers, to carry them along a path analogous to that already
trodden by the neo-Platonists, toward asserting the utter simplicity of God.
The Hadith folk objected that this was to empty the notion of 'God' of
all content — and was hardly better than denying God altogether. Al-Ash'ari
seems to have insisted that God did have attributes, and to have met the
Mu'tazilis' point by saying, neatly, that they were not other than His essence.
The logical categories were retained, but were not allowed to detract from
the effectiveness of a God that could intervene in history.

He might have liked to avoid the Philosophers' categories altogether. We

have noted how the notion of tmdn, 'faith' in God and His message, was

intellectualized into sheer 'belief, assent to propositions, especially among

the Mu'tazilis. The Hadith folk, as conscious of needing to define the godly

community as the Mu'tazilis, agreed in principle; but they felt that faith

must require a more personal involvement as well, citing Qur'anic passages

11 J. Schacht, 'New Sources for the History of Muhammadan Theology', Studia
Islamica, i (1953), 23-42, has a variety of useful observations on the relation of the
schools of kalam to other Islamic currents. (Despite the title, it is not about a theology
of Muhammad or one derived directly from him personally, but is about Islamic theology
in general.)

440 THE CIVILIZATION OF THE HIGH CALIPHATE

in which faith was said to increase in quantity — which sheer assent could
hardly do. The expressive anecdotes of hadith and the Sirah, the life of the
Prophet, best set forth what was required. But kalam disputation required
a concise general formulation. In this case, al-Ash'ari did not attempt a new
sort of abstraction from the texts, nor even a nice logical distinction. He
turned to formulations of 'faith' that had been extracted from the texts for
legal (not doctrinal) purposes, for the sake of defining who is entitled to the
rights of a mu'min, a man of iman, faith, by his hadith-minded friends.
'Faith', then, included assent and affirmation and action as well, though the
assent was primary. In this way he introduced into systematic kalam a
formulation which made the essential point, yet compromised as little as
possible the human quality of the texts.

By the latter part of the tenth century, several schools of kalam dis-
putation were actively combating each other, sometimes violently. Besides
the still vigorous Mu'tazilis and the Sufi-inclined Karramis, the two most
important schools traced themselves to al-Ash'ari and to al-Maturidi,
respectively. The Ash'arts were associated with the Shafi'i legists, the
Maturidis (and the Mu'tazilis) with the Hanafis. The Ash'aris went the
furthest toward a strict adherence to the positions of the Hadith folk, and
developed withal a highly sophisticated interpretation of cosmology which
moved as far as logically possible from the positions of Falsafah.

The character of the cosmology of the Muslim Faylasufs becomes especially
clear when contrasted to the cosmology of the Ash'ari kalam. The Faylasufs
were interested, since the days of Plato, in the unchanging, in the permanently
valid. Thrust into the water, a stick appears bent; in the air, it appears
straight. When one is angry, one's neighbour seems an object for violent
assault; a few minutes later, he may seem an object for pity. If one is born
in India, it seems of the utmost importance to burn one's father's corpse; if
one is born in Arabia, one will bury it, and do one's best to prevent anyone's
burning it. A year ago one's fields were rich with wheat and this year the
same fields are almost barren. In such a world what can one be sure of?
The rationalistic answer of the Philosophers was that though individual
plants and even fields appear and disappear, we can know what wheat is, as
such, and what a field is, and what is universally true of any wheat growing
in any field; we can know what anger is, and what pity is, and what a human
being is as such, apart from any particular feeling we may have for particular
persons. Knowledge is therefore a matter of timeless concepts, essences, and
natural laws, rather than of transient and changing details. We can be
sure that there are 180 degrees in a triangle, that justice is more admirable
in men than injustice, that oaks grow from acorns ; we cannot be sure, but
can only have a provisional opinion, that this three-cornered piece of wood
is a triangle, that this man is just, that this acorn will actually grow into
an oak.

The Ash'aris, on the contrary, doubted that there were any inherently

speculation: falsafah and kalam 441

unchanging essences and natural laws. For them the most important facts
were not abstractly universal but very concrete and historical. These were,
first, that the individual man Muhammad had brought to human beings
supreme truth in a particular place at a particular time, and that this truth
was carried by his community from generation to generation; and second,
that every individual was faced with the supreme choice of deciding in his
own case whether to accept this truth or not. One could know the individual
man Muhammad, or more exactly one could know by documented hadith
reports, various individual facts about him; it was much harder to say
anything dependable about the universal essence of prophecy. The Ash'aris
granted that certain sequences of events tended to repeat themselves — God,
as it were, has habitual ways of ordering nature. But they insisted that we
have no evidence that events must necessarily always repeat themselves;
all we can actually know is the concrete momentary fact.

The Faylasufs, looking to unchanging essences and laws, thought of a
world tunelessly proceeding from self-sufficient Reason, each event in it
being but an exemplification of logical possibilities. Mankind might change
its condition from time to time, societies rise and fall and learn and unlearn;
but there could be nothing inherently new. Such men as learned to fulfill
the moral and intellectual demands of their truer natures might purify their
intellectual spirits to the point that after death they were released into the
realm of the changeless. The wise man was summoned to a virtuous con-
templation and knowledge of existence, from the objects of the individual
sciences up to the nature of Divine Reason itself.

The Faylasufs saw a realm of natural order transcending the arbitrary
power of isolated and unpredictable events, which power had formed the
basis for the religious awe in which unthinking persons tended to hold
everything that was out of the ordinary. But such a realm of nature seemed
to allow meaning only to the type — even the human being, as an individual,
could be meaningful only so far as he came to embody an abstractly rational
nature. Monotheists of the prophetic tradition found themselves too deeply
and individually challenged by personal moral demands to accept such a
world. In God they saw a power that transcended in turn the order of nature
— and thereby made possible meaningful human existence apart from mere
fulfillment of natural types. The Ash'aris wanted to safeguard recognition of
that divine transcendence at every cost (even if it seemed incompatible with
the notion of free will in a person, which in any case could be no more
compromised than it was by the Faylasufs' determining rule of cause and
effect) — for only in God's presence could they feel that they possessed, on
whatever terms, a personal responsibility.

Accordingly, the Ash'aris, looking to revelatory events and authoritative
hadith reports, thought of a world produced as an act of will in time, as an
event itself, by the Divine sovereignty; and within it, every particular event
was in turn the immediate act of God. Mankind had had a beginning and

442 THE CIVILIZATION OF THE HIGH CALIPHATE

was having an all-important history, in the course of which God gathered
peoples to His obedience through His prophets. Such persons as obeyed
God's commands would be rewarded at the end of history in Paradise, and
the others punished in Hell. For the Faylasiifs, it was the wise and knowing
elite, the few who could become objective scientists and philosophers, that
were the only persons really fully human. For the Ash'aris, on the contrary
(sure that a little learning is a dangerous thing), the Divine blessing was
upon the average individual whose overt acts conformed with the pattern
laid down for the community of Muhammad.

Kal&tn as a system: (he problem of divine power

On this basis, the Ash'aris — by the end of the tenth century — developed a
comprehensive system. A central problem with which they started was that
of the relation of God's power to human actions. The Mu'tazili solution
had usually been too simplistic. Al-Ash'ari himself is said to have posed to
his Mu'tazili teacher (the great al-Jubba'i), who frankly asserted that God
always rewarded humans as they deserved since humans could choose their
actions at will, the question of the three brothers : One brother had a high
post in Paradise, having lived long and done many good works, and the
second brother, with a lower post, complained that he had died young and
had not had the opportunity to do as much good as his brother; whereupon
God explained that he had foreseen that the less-rewarded brother would
have sinned had he been allowed to live longer, and would have been still
worse off. At that, the third brother raised his voice from Hell, demanding
why God had not cut him off early too.

Clearly, there was no way, within the bounds of ordinary logic, to declare
God at once omnipotent and omnibenevolent. It was more realistic just to
say that all was His will, without explanation or justification. Yet at the
same time, it would not do simply to ascribe every movement that happened
to God: one had to make some sort of distinction between the actions of a
responsible human being and the motions of a stone as it falls. The solution
to this problem advanced by the Ash'aris was an attempt at simple des-
cription of the moral situation. First they pointed out that good and evil,
indeed the laws of logic itself, are what God decrees them to be and it is
presumptuous for human beings to attempt to judge God (or to justify
Him!) on the basis of categories God has laid down only for human beings.
In particular, what humans were to be held responsible for was God's
arbitrary decision. Hence human responsibility was not, as some Mu'tazilis
had said, a function of the way that the action is produced — by choice as
against inherent nature; rather, God alone creates all actions directly, but
in some actions there is superadded a special quality of voluntary acquisition
(kasb) which by God's will makes the individual a voluntary agent and
responsible.

speculation: falsafah and kalam 443

But a more theoretical solution of the problem, at the same time an
expression of a fundamental monotheistic viewpoint on nature and history,
lay in the theory of existence which the Ash'aris worked out. The world
was not made up of entities, enduring substances with their accidents, each
with its own nature which accounted for its activities and its effects. This
was to introduce secondary causes besides the immediate willing of God.
Rather, the world was made up of atomic points, each at a given moment
in space and time, among which the only continuity was the will of God,
which created every atomic point anew every moment. If a given object
seems to endure over a period of time, it is that God creates at every moment
a new set of atomic points corresponding to what was there. Certain sequences
and combinations of atomic points answer to God's customary ways; but
none is necessary.

In such a world, of course, there can be no such thing as miracle: there
can only be wonders, breaches in the normal custom of God, which, however,
are in themselves just as 'natural' as any other event; or, rather, everything
is a miracle, a special intervention of God's power. (Accordingly, the Ash'aris
could allow for any amount of miraculous accounts of the prophets, but
at the same time avoided laying any great weight on such mere wonders;
for evidentiary miracles, wonders which were to prove anything, they
demanded a clear and declared connection with a divine message — so that
it was, finally, the message which proved the miracle, rather than the
reverse. This effectively put the emphasis, in recognizing a prophet, on the
quality of his message rather than on any extraneous signs.) In such a
world, also, human actions could not be ascribed any organic internal cause
and effect ; they were as much the creatures of God as any other. Yet God
could give any human action whatever psychic quality He willed. Finally,
in such a world, the only dependable knowledge was historical knowledge —
and knowledge of the truth revealed in historical moments: in any case,
knowledge of explicit individual events. All generalization, however useful
it might be, was hazardous; and such generalization could never, in any
case, lead to ultimate truths about the nature of life and the universe.

Such kalam disputation was militantly (and, one might add, philosophic-
ally) anti-Philosophical. Nevertheless, kalam of every school was long
suspect in the eyes of many, perhaps even most, scholars of fiqh law, and of
the more consistent of the Shari'ah-minded generally. Till well after High
Caliphal times, its position was at best secondary to that of the Shari'ah
in the eyes of most of the 'ulama.'. If it was true that speculative generaliza-
tion could in fact, as this atomistic analysis suggested, lead nowhere, then
why speculate even so much?

Adab: The Bloom of Arabic

Literary Culture

c. 813-945-

If the Shari'ah-minded, and generally people of a pronounced Muslim piety,
tended to be chary of associating with the court, and if even the Faylasuf
scholars, who often depended personally on court patronage, commonly
rather despised its ways, the court continued nevertheless to sit at the head
of the Islamicate society as a whole. It was there that the most striking of
the decisions were made which set the political, social, and even, largely, the
economic context which all other currents of the high culture presupposed
in practice. On the central government and its absolutism depended the
peace and prosperity within which all the new cultural flowering was taking
place.

But this absolutism depended not on sheer military might but on cultural
expectations which could legitimize it. That is, it required habits of mind
that would lead people to look to its restoration if it fell on hard times, rather
than merely to enjoy or suffer its strength when it was strong and look to
whatever seemed likely to take its place if it faltered. Only so could it have
a chance to recover from serious temporary defeats. For maintaining such
habits of mind, the absolutism could depend but little on the grudging
recognition afforded it by the Shar'i 'ulama' It must depend heavily on the
attitudes of a large class of officials and bureaucrats, together with wider
strata of landowners and even rich merchants from whom the officials were
drawn. Such men placed their greatest political hopes not in a problematic
universal justice (through which they, indeed, might have been levelled
down into the masses), but in a more realistic system which should assure
the good order necessary for the security of those who were fortunately
placed — and for the highly cultured living which for them constituted the
good life.

The katibs, administrative clerks, seem to have come to differ, under the
High Caliphate, from their earlier equivalents, even those of earlier Islamic
times and presumably especially from those in Sa.sa.nian times, in being
drawn less exclusively from the old Iranian gentry, with their local roots in
an agrarian order, and more freely from Arab families or nearly risen converts

444

THE BLOOM OF ARABIC LITERARY CULTURE 445

or from mercantile and other urban elements, which had little in common
save their bureaucratic work and well-paid social standing. Such an
administrative and cultural elite was not only city-based but cosmopolitan
in outlook. They acknowledged the ideals of the 'ulama.' scholars but did
not take them too seriously, and they regarded the ideals of the Faylasufs
as an esoteric specialty for a few. Their culture can be summed up under the
heading adab, the pattern of cultivated living which grew up around the
court and in the provincial centres and was imitated yet more widely. It
was in this culture that must be found those enduring patterns of expectation
which could give solid support for the absolutist tradition, independent of
the limitations imposed on it by Shari'ah-mindedness,

The adab culture which centred on the court had its own serious triumphs,
which gave it more than a transient dignity. It was at the court that the
fashion was set, and the highest level of achievement displayed, in important
intellectual and aesthetic dimensions of culture which Shar'l 'ulama' and
Faylasufs, as such, almost ignored. By the time of the caliph al-Ma'mun at
the start of the ninth century, these patterns too were being given a new
classical form which was being accepted, as was the new religious allegiance,
throughout the Muslim domains. All more limited older traditions of elegant
living faded in its presence, even that which had prevailed at the Sasanian
court. This new cultural tradition bore a large part of the burden of assuring
the prestige of the government and with it the stability of the whole society.

To this extent, the ultimate failure of the absolutism to maintain itself
can be traced in the failure of the adab culture to establish a permanent basis
for it. The prestige of the caliphate allowed a certain time of grace during
which a basis might be built. But by al-Ma'mun's reign, this time was
beginning to run out. The adab culture laboured under several handicaps.
First of all, the basis had to be built without serious reference to any cultural
norm of any longer standing than the Muslim communal allegiance itself.

The breach with the older traditions

In Europe, in India, and in the Confucian Far East, the classical languages
and the cultural ideals of the Axial Age continued to be studied directly and
to influence fundamentally the regional high-cultural life down to modern
times. But in the Nile-to-Oxus region there had been no single classical
lettered tradition since Cuneiform times. Each confessional community had
had its own; the Muslim community likewise had established its own
tradition. This grew out of the older traditions but looked to its own creative
moments which had been experienced in the new language and within the
new religious allegiance. The central elements in the Islamicate cultural
background remained (as in the other core areas) those of its own home ground
— the Irano-Semitic lands; but the actual documents of the ancient Irano-
Semitic cultures were no longer studied. Not merely were Cuneiform classics

446

THE CIVILIZATION OF THE HIGH CALIPHATE

such as the Gilgamesh epic quite forgotten; even the masterpieces of the
prophetic traditions were — if not forgotten — at least neglected. For the
cultural achievements of even the Axial Age and since were for the most
part retained only in a drastically Islamized dress.

The Earlier Classical Arabic Belles-Lettrists

Fifth and Pre-Islamic Arabic poetry — the 'classical' qasidah ('ode') style;

sixth the Arabian poets Imru'1-Qais, Tarafah, Zuhair, and others;

centuries important Sasanian and Byzantine cultural eras also

Early eighth Umayyad poets Jarir, al-Farazdaq, and al-Akhtal; rise of ghazal
century (love song, and generally lyric) style — 'Umar b. 'Abi Rabi'ah,

d. 719

c. 760 Death of Ibn-al-Muqaffa', translator of Persian works, caliphal

adviser, and prose writer
c. 815 Death of Abu-Nuwas, court poet of new styles, detractor of the

old poetry

c. 828 Death of al-Asma'i, Arabic grammarian and lexicographer,

collector of old Arabic poetry

c. 845 Death of Abu-Tammam, collector of old Arabic poetry, poet in

his own right who imitated the old style
869 Death of al-Jahiz, Mu'tazili theologian, master of the Arabic

prose essay

889 Death of Ibn-Qutaybah, grammarian, theological and literary

critic, prose writer in the spirit of adab, moderate exponent of

new forms
892 Death of al-Baladhuri, collector of hadiths; wrote history of

Arab conquests
923 Death of al-Tabarf, master exegete and hadith-based historian

of pre-Islamic and Islamic periods
951 Death of al-Istakhri, geographer, who wrote a description of the

world using work of al-Balkhi (d. 934)
956 Death of al-Mas'udi, well-travelled and erudite writer,

'philosophical' historian
965 Death of al-Mutanabbi', last great poet in older Arabic style,

paragon of subtlety in poetic allusion

This breach in continuity made by Islam answered (on a more total scale)
to such breaches made by all the lettered traditions between Nile and Oxus
almost from the time when the Cuneiform tradition was abandoned: in its
monotheistic exclusivity, each new religious community tended to reject the
culture of outsiders as false. Perhaps among some Muslims there was a
certain hostility to agrarian aristocratic tradition in the rejection of at least
Pahlavi Sasanian culture. But the Muslims were able to make a more
effective overall breach in cultural continuity than had previous groups,
ultimately because of the remarkable degree to which the more populistic

THE BLOOM OF ARABIC LITERARY CULTURE 447

and moralistic of the monotheistic religious aspirations were fulfilled in
Islam, especially in its Shari'ah-minded form. It was, initially, the effort to
fulfill these religious aspirations that set off the Islamicate society from its
past and, in effect, assured the distinct existence of an Islamicate civilization
at all. Hence these aspirations had a key position in withholding or acknowl-
edging legitimacy to any other tradition in the civilization as a whole, even
though they may have done little more substantively than colour the content
of that civilization. To such withholding of legitimacy, a lettered tradition
was singularly vulnerable, being carried anyway by a minority and subject
to the sway of opinion. It was a strong sense of the religious alienness of the
Greek, Syriac, and Pahlavi traditions that prevented their classics from ever
achieving a legitimate status as cultural authorities in the new community.
It was just in fulfilling the communalistic ideals of the most active of the
older traditions that Islam built up the exclusivity that found those traditions
alien.

The most immediate victims of the Islamic exclusivity were the other
Abrahamic religious traditions. Already Muhammad had accused the
Christians and Jews of having misinterpreted the messages sent to them.
At the least, according to the Qur'an, the Christians had of their own accord
invented added obligations, notably monasticism, which God had never
imposed on them; much worse, they (or many of them) had misinterpreted
Jesus and Mary, making them objects of worship alongside God. The Jews
had likewise brought upon themselves added burdens because of their
chronic hard-heartedness, and those of Muhammad's own time had further-
more suppressed scriptural evidence that would have supported Muhammad's
mission, out of jealousy that any but themselves should be divinely favoured.
The Muslims were not to have fellowship with such secret enemies. Yet
nothing in the Qur'an required the total rejection of their books.

Among the early Piety-minded, however, these indications were developed
into a comprehensive condemnation of the two peoples of the Bible generally,
not merely their representatives in Muhammad's Arabia. First, they had
corrupted their own scriptures by suppressing some passages and distorting
or interpolating others. Second, their learned men had nonetheless been
able to know (from these same holy books) that Muhammad was to come,
and had recognized the signs of his coming, yet out of jealousy the Christians
and Jews had refused to admit his prophethood. Consequently the peoples of
the Bible, having wilfully rejected Muhammad and the Qur'an at that time,
and having generation after generation persisted in their contumacy, were
not merely ignorant but guilty in their refusal of Islam; if they were
tolerated, it was not as fellow-worshippers who happened to be partly
mistaken, but simply by command of God in His unfathomable mercy.
Correspondingly, their books, far from being mines of information about the
earlier prophets whom every Muslim was pledged to acknowledge, were to
be shunned as worse than merely human books, for such true revelations as

448 THE CIVILIZATION OF THE HIGH CALIPHATE

might be among them were mixed with impious falsehoods inspired by the
Devil himself. Anything of value in their books would have been brought to
Muslims by early converts, whose reports about Torah and Gospel, trans-
mitted as hadith, were alone to be trusted.

Zealous Shari'ah-minded Msulims elaborated gladly their code of symbolic
restrictions on the dhimmi non-Muslims— they must wear certain humiliating
garments and signs, they must not build new houses of worship, and so on
— most of which was retroactively ascribed to 'Umar; later caliphs such as
al-Mutawakkil made some effort at imposing it in practice. As the dhimmi
communities dwindled, popular Muslim sentiment more and more readily
insisted on stigmatizing the minorities. But one of the most fateful restric-
tions was that on the Muslims themselves : unlike the Christians, who could
despise the Jews yet read the Hebrew Bible, the Muslims, more consistently
communalistic, were debarred from the Bible altogether. In High Caliphal
times, a certain number of intellectual Muslims, finding themselves in close
contact with Christian colleagues, were able to develop a serious critique of
Christian beliefs, reading parts of the New Testament in translation. But
such efforts remained marginal and had little reflection in the overall
development of Islam.

Accordingly, even so explicitly honoured a body of classics as the Bible —
the corpus embodying the Hebrew prophetic tradition as well as its Christian
development — was known to the Muslims chiefly in the form of corrupt and
often legendary fragments. It exercised its influence more through the
unconscious continuity of social patterns than through even an indirect
literary tradition — let alone a direct confrontation of each new generation
with the records of the original inspiration.

In fields where religion was less explicitly at stake, the discontinuity
could be less drastic. But even the most important of those traditions never
gained a universal standing. In the culture of the adib, apart from Islam
itself, four chief traditions played a conscious role. The adib was fully
conscious of a certain Arabism, associated with the nobility of the Mudari
Arabic language and of the pure camel nomads, its titular carriers. With this
was also associated, in theory, the old high culture of the Yemen; but in
practice, nothing of that survived save by way of peripheral legends. The
adib was also aware of an Iranianism, associated with the glory of the
Sasanian court and of the wise emperor Nushirvan, in whose time Muhammad
was born. This awareness had a major impact. He rejected explicitly the
cited Semitic tradition carried in the various Aramaic tongues and associated
with underlings even in Sasanian days — yet many of the Arabic lettered
patterns, religious and otherwise, were tacitly taken over from Aramaic.
Finally, he was ambivalent to the Hellenic tradition, associated with the
sages of antiquity but also (since Christian days) with paganism.

Greek science and philosophy maintained {in translation) such universality,
at least, as they had previously achieved between Nile and Oxus; the Muslim

THE BLOOM OF ARABIC LITERARY CULTURE 449

Faylasufs never entirely replaced the older writers as points of departure for
the continuing tradition and its dialogue. But that tradition was not the
overall Hellenic one and even in its truncated form it was less at home in
Islam as a religious community than it had been even in some of the previous
communities; it could never form the primary intellectual impulse in
Islamicate culture.

The heart of the pre-Islamic heritage of Muslim Arab culture lay, in many
respects, in the Semitic Aramaic traditions of the Fertile Crescent. But this
was true especially of the several aspects of the religious tradition, since
Islam was a development of the Abrahamic prophetic tradition; and it was
just in the sphere of religion that direct reference to the older heritages was
suppressed. The secular literature of the Aramaic languages, except so far
as it embodied the tradition of Philosophia, may have been relatively
unimportant as compared with the communally religious literature. It
served as a model — how extensively, it is hard to judge on the basis of the
surviving Aramaic and Syriac literature, largely religious ; but it was despised
by the privileged circles as the heritage of peasants. (Yet persistent tradition
made it the language of Adam and the earlier prophets.)

Pahlavi, however, had been becoming a rich cultural language in all
dimensions at the hands of the Sasanian ruling classes. The communally
religious books of the Mazdeans were despised and shunned even more
readily than those of Christians and Jews (though even this condemnation
was not made inevitable by the Qur'an) ; they were felt to be obviously false
and he who even looked at them was suspected of betraying Islam. But the
works of history and of belles-lettres, as well as of natural science, could not
be so readily condemned. Moreover, it was this secular Pahlavi tradition
that had embodied the cultural support and legitimation for the absolutism
of the past. In the name of the caliphal state itself, some Muslims had
appealed to its cultural standards as socially indispensable as well as humanly
insurpassable. Here the Muslim exclusivity worked more subtly.

A conscious movement, which did arise, to depreciate outright the Arabian
elements in Muslim culture, had only limited success (at least at the capital) ;
in fact, in the Fertile Crescent, as well as in the formerly non-Arabic southern
fringes of Arabia and even in Egypt, large areas were on their way to adopting
versions of Arabic as the popular family language; and with the Arabic
language came a pride in the Arabian heritage coupled with scorn for those
(generally more rural and backward) elements that retained the Aramaic
tongue and were scornfully called 'Nabataeans'. But elsewhere, the ordinary
family language remained non-Arabic and the ordinary people, on becoming
Muslims, could not closely identify themselves with the Arabs as such. There,
cultural theory was less relevant, positively or negatively, than the reality of
everyday life.

In the Iranian highlands, in particular, Iranian tongues were the rule and
in the greater part of the plateau the various dialects were closely enough

450 THE CIVILIZATION OF THE HIGH CALIPHATE

related to justify a single common Persian literary language accessible to all.
This had been Pahlavi. But literary Pahlavi had been used especially by
priests and clerks of the bureaucracy, the most important literate elements
in the more agrarian, Iranian parts of the Sasanian empire; and it had been
written in an uncommonly clumsy manner, which laymen such as merchants
must have found exasperating. Moreover, the expectation had grown up that
each religious community, along with its own literary language, would use
its own special script, hallowed as it was by association with the community's
sacred writings. When Persian-speaking Muslims who had been trained not
in Pahlavi but in Arabic turned to using the Arabic script rather than the
Pahlavi for their own Persian, the idea rapidly caught on as both pious and
convenient. By the end of the High Caliphate, when Persian was written by
Muslims it was being written not as Pahlavi but in a new form, expressed in
Arabic characters and, incidentally, reflecting such changes in pronunciation
and grammar (not very extensive) as had supervened in the spoken
tongue since the standardization of literary Pahlavi. Especially in Khurasan,
under a dynasty of governors of Persian origin (the Sa.ma.nids), this new
Persian was being erected into a literary tongue, especially for poetry.

But the result was that even among Persians the old Pahlavi came to
require special training. Pahlavi literature continued to be read by some
Persian Muslims throughout the High Caliphal period, but the difficulty of
learning it raised complaints. And while a certain amount of Pahlavi was
translated into not only Arabic but, later, the new Persian, yet the new
Persian had its own traditions tinged with a Muslim atmosphere, into which
the Pahlavi materials scarcely fitted without a certain amount of adap-
tation. Thus the Muslim Persians cut themselves off linguistically from the
Pahlavi past and adopted literary standards more closely bound to those
of Muslim Arabic than to those of pre-Muslim Persian.

The Pahlavi tradition, in contrast to the Greek, did inspire a large part
of the Islamicate lettered traditions in fields which received general recog-
nition, most specifically in courtly and literary adab; thus it substantially
extended the range of populations under its influence. But it did so not in its
own form but in new guises. No more than the Greek did the Pahlavi, in
original or in translation, establish a strong enough position so that it would
be assumed that a learned man would be acquainted with its classics; and
unlike Greek, Pahlavi, even among the Persians (apart from a few adaptive
translations), did not long survive the High Caliphate, except for religious
purposes among the Mazdeans themselves.

The consequence of all this was that, in contrast to Europe, China, and even
India (where such works as the Vedas and Upanishads do breathe a very
different spirit from that of Shaivism and Vaishnavism), in Islamdom there
came to be little direct contact with the great human works of the pre-
confessional periods in the Axial Age. The humanistic insight available in a
work such as the story of David in the Hebrew Books of Kings was excluded

THE BLOOM OF ARABIC LITERARY CULTURE 451

in favour of classics all of which presupposed the Muslim allegiance. Apart
from the almost esoteric tradition of the Faylasufs, there was no heritage of
classics transcending that of the community and its revealed origin. (The
pre-Islamic Arabian poetry, which was retained, could not seriously fulfill
such a role any more than the Niebelungenlied could have done so for a
Germanic culture which rejected Latin.) The communal tendency of populism
in the region had come to full flower. This characteristic of the Islamicate
civilization, unique in the Oikoumene, perhaps made for a relative spiritual
impoverishment of it as compared with the major civilizations contemporary
with it. But it could also make for a uniquely self-sufficient strength and
cultural integrity — a strength which, in subsequent centuries, was to play
its part in allowing Islamdom to expand all over the hemisphere without
great danger either of losing local cultural roots or of destroying the solidarity
among all members of a far-flung community. 1

The cultivation of an adib

Meanwhile, the cultural resources of Islamdom were brought together and
focused in the adab culture of the court. The adib, the possessor of adab,
was the man of a varied and brilliant set of attainments which were the
adornment of his society. With public order assured by a strong monarchy,
the private good of the privileged who benefited thereby was a life of good
taste.

If one can distinguish two poles between which range our criteria for
judging the quality of human life, a pole at which moral relationships among
people take precedence over all else, and a pole at which what matters is the
excellence of learning and art and the fine use and appreciation of human
resources, in however few out of a given population, then the Shari'ah-
minded stood at the moralistic pole and the adibs very nearly at the other.
For them the justification of a man's life was at least in part the degree to
which it was cultivated and refined and could with beauty make use of
wealth and leisure. A Muslim rarely said as much explicitly, but such a
gentlemanly ideal had been implicit in Aristotle's ethics, and it was Aristotle's
ethics, out of all the corpus of Falsafah, that gained the greatest vogue—
in appropriate Arabic adaptations — among the adibs. While the 'ulama'

1 On the process of transition of the high culture from earlier languages into Arabic
we have no adequate studies, largely for want of adequate knowledge of the state of
the several traditions in Sasanian times. We do have a number of studies on the transition
from Greek (not limited to natural science and metaphysics), notably those by Becker,
von Grunebaum, and Kraemer already noted in footnote 8 to the General Prologue.
Rudi Paret has also written on this and is always worth consulting: Der Islam und das
griechisphe Bildungsgut (Tubingen, 1950). (I might add that the most accessible study
in English, De Lacy OXeary, How Greek Science Passed to the Arabs [London, 1949], like
others of O'Leary's works, is marred by errors of fact and of judgment, and does not
add to the more standard authors on Arabic science.)

452 THE CIVILIZATION OF THE HIGH CALIPHATE

scholars looked to the moralistic and populistic strains in the Irano-Semitic
background and refonted them, the adibs looked to just the opposite
sorts of strains in that background and produced a new culture in their
spirit.

But the adab of caliphal times seems to have diverged from that of
Sasanian times not only in language but in spirit. It was evidently — to
judge largely by the older fiction that has been preserved — less rooted in
local legends than the Sasanian, more city-based, more cosmopolitan. The
idealists among the katibs, following older models, tried to link adab culture
to birth and breeding, and taught each other to despise the upstart merchant.
But this was to fly in the face of the realities of social mobility. Adab could
no longer presuppose the homogeneity of the Sasanian gentry. Nor did it
produce a new homogeneity similarly rooted in the land and devoted to the
caliphal power.

Adab was necessarily, to a certain extent, the way of a particular class,
that of the administrators and officials who depended ultimately on the
caliphal authority. The special professions of the adib were those of secretary
and clerk, of bureaucratic manager. But adab was cultivated eagerly by
well-to-do merchants and by anyone who wished to be up with the times.
Adab cultivation entailed, in principle, a comprehensive synthesis of all
high culture. It made a certain place for the Shari'ah-minded learning and
its requirements, and gave a certain honour to the 'ulama' ; in turn, some of
the 'ulama' made a point of cultivating the social graces and even appeared at
court. Some touch of knowledge of Falsafah was an asset to an adib. But adab
had its own unmistakable spirit.

A central position in adab was held by Arabic literature: especially the
poetry and the rhymed prose, which held a position of honour not accorded
to any other arts. The spoken word well put moved cultivated men as nothing
else in life was permitted to, and its refinements were explored by a galaxy of
remarkable verbalists. Above all, then, the adib should have a skilled
command of the standard Mudari Arabic language. Good speaking, as well
as good writing, was the supreme mark of good breeding and of polished and
enjoyable intercourse among men. This meant, in the first place, knowledge
of the grammatical intricacies of Mudari Arabic — a desideratum not only
for non-Arabs but for Arabs themselves, whose vernacular was already
shifting away from the Arabic of the Qur'an; second, command of its rich
resources of vocabulary. For every sort of composition, and preferably also
for conversation, a linguistic brilliance, in which the right thought was put
in the cleverest possible way, was admired; it decorated the scene as much as
did rich robes or flowered gardens. It is for this reason that the discipline of
grammar, early developed among the Piety-minded in Kufah and especially
Basrah, took on such a great social importance as part of the equipment of a
man of the world.

But in addition to a sheer command of language, the adib should know a

THE BLOOM OF ARABIC LITERARY CULTURE 453

bit about everything of interest to the curious. The ideal state letter, however
prosaic its content, should be an elegant production. Gradually standards
became ever more fully developed, as the styles of certain individuals won
admiration and became models to be imitated and then surpassed. Finally,
it was taken for granted that every serious missive would be adorned with
rhyming prose and resonant periods. Any composition of importance should
be varied with poetic quotations and illustrated with references to obscure
points of learning. Piety and sobriety of living were highly valued by most
adibs ; yet an elegant style of life (rather in the sense used in 'styling' Ameri-
can automobiles) and especially an elegant style of verbal production were
given great honour. Any man who could show himself capable of a rounded
and high-level adab was assured of good opportunities in the court society.

What an adib should know: history and geography

Ideally, all kinds of studies should nourish a rounded adab culture. But
distinctions were made. Literary studies in Arabic and studies associated with
the Shari'ah law (including history) were called 'traditional', 'transmitted',
studies, for they depended on historical information about events and
conditions which would not recur (even Arabic grammar, for instance, was
based on the speech of the Arabs of Muhammad's time and no other) ; some-
times they were called 'Arabic' studies because of the special role knowledge
of Arabic played in them all. They were contrasted to 'rational' or 'non-
Arabic' studies, such as natural science, which could (in principle) be
developed from scratch on the basis of experience available at any time; and
which were in fact known to have been cultivated also in the earlier lan-
guages of culture. It was the 'Arabic' studies that naturally had the highest
prestige among the adibs, as much for literary as for religious reasons.

The pious adib, then, should know as much as possible of the Shar'i
fiqh as it was then being worked out. Knowledge of fiqh was practical both
for a private individual and for a state clerk, for it was the officially recognized
basis of social order; knowledge of it also implied piety. At the same time,
he should have a wide acquaintance with history (and geography) ; and he
should know the famous tales and sayings to which allusions might be
made. Naturally, he should be familiar especially with courtly precedents.
Finally, he should command something of the natural sciences. But always
he should have a good knowledge of poetry. This should include both know-
ledge of the rules of poetic propriety that were being built up by the gram-
marians, and the memorizing of as many lines of verse as possible, to be
produced on appropriate occasion. Anthologies, encyclopedic surveys of
learning, and descriptive catalogues of available books helped the adib keep
abreast of it all.

The learning of the adib, whatever the field, was never clearly distinguished
from his concern with belles-lettres. When the adib studied 'biology', for

454 THE CIVILIZATION OF THE HIGH CALIPHATE

instance, he was not so much interested in learning the structure of organisms
as in finding out all the strange things that could be said of them. A dis-
course on animals commonly reads more like Ripley's Believe It or Not than
like a text in zoology. The aesthetic propriety of various ways of making
literary references to the animals discussed was a major concern. All know-
ledge was a means of adorning and enriching literature.

Historical writing — from which could be drawn examples piquant or
cogent to enliven every point — was popular in a range of forms which
illustrate the several strands that entered the adib's thinking. Adibs read
history in the several genres used by the 'ulama.'. The most important of
these was the annalistic chronicle — a form carried forward from Syriac
models, which might be modified (as by Tabari) with the addition of the
isnad documentation and an Islamic orientation. Perhaps less important
for adibs were forms serving special Shar'i needs : the biographical collections
which ranged prominent Muslims according to the generations elapsing
since that of Muhammad's associates, or the histories of dogma which
recorded formulations of doctrinal position. But other sorts of history,
wherein the literary effect or the courtly orientation were more prominent,
were more closely associated with adab. For instance, the stories of the
battle days of the Arabs: commonly joined with bits of poetry referring to
the event, they had been preserved in an oral tradition and (written down)
entered into the mental world of the adib as much in connection with poetry
as with history proper. To the extent that this old Arabian history had any
continuity, it was not, of course, in terms of the chronology of the community
of Muhammad, like the Shari'ah-minded history, but in terms of tribal
genealogies.

This anecdotal history, stemming from the old Arabian background, was
of relatively minor importance. In an urban setting it was the history of
kings that loomed most impressively. Greek history (geographically marginal
to the area) seems to have been very little known except as it came deformed
through late Christian sources. The Hebrew Biblical historical tradition
remote in time, likewise was known chiefly by way of debased oral reports
from such people as converted Jews. Much more imposing was the Sasanian
history, which was in large part transferred into the new court language,
Arabic. It centred neither on the development of the community nor on
tribal raids and battles, but rather on the reigns of hero-kings. These were
presented as enormous figures embodying in themselves the whole social
order, in accordance with the old Iranian conception of absolute monarchy.
A good monarch produced a blessed age; a weak or evil monarch meant
catastrophe. Along with other elements of Sasanian prose, this historical
tradition entered not only Arabic learning and literature but, through the
Muslim Persian language, all later Islamicate literatures.

Sasanian historical attitudes could pervade historical materials from other
sources. The Sasanian tradition of Iran and the Biblical tradition of the

THE BLOOM OF ARABIC LITERARY CULTURE 455

Fertile Crescent were uneasily co-ordinated to form the main body of pre-
Islamic history. The first Iranian man was identified with Adam, who could
take on kingly traits in consequence; and Solomon (Sulayman) became an
emperor as great as those of the legendary Iran. But the Sasanian viewpoint
was only one among several elements in historical writing.

Out of these backgrounds developed a variety of Islamicate historical
approaches suitable to the special needs of the adib. Even in historical
material in which a courtly approach predominated, the isnad documen-
tation favoured by the 'ulama' might appear. Several ways of studying
history were represented in the careful work of al-Baladhuri (d. c. 892),
who studied Muslim history from a more secular viewpoint than his younger
contemporary, Tabari. He wrote a massive history of the Arab conquests,
arranged as a series of monographs on each area conquered, and including
a number of documents verbatim or abridged. In connection with the
conquest, he brought in a good deal of local administrative information —
especially appropriate, since the local tax and administrative status was
often supposed to depend in part on capitulation terms made at the conquest.
He also made a comprehensive study of the chief Muslim families — arranging
it in genealogical form (like some others who specialized in genealogy),
but giving the sort of extensive biographical information that others were
arranging by generations from Muhammad — and also including much general
information, not neglecting central administrative detail, when he came to
deal with individuals who served as caliphs. This arrangement by families
was especially appropriate to those who set store on ancestry and looked to
family tradition as a major spur to loyalty and high standards in the
descendants.

'Ali al-Mas'udi (d. 956) represents more obviously the desire for bits of
curious information from all over and combines with his historical tidbits,
organized by reigns and chronologically, surveys of geography, astronomy,
and all the rest of the world of unlimited remarkable facts on which the
adib could draw to ornament his conversation and his state letters. He
represented not only literary spice, however; his works embodied a philo-
sophical view of mankind which allowed him to transcend communal lines in
his curiosity and in his appreciation, in both time and space.

Geography was frequently treated as an independent discipline. It was
useful for administrative (and commercial) purposes. It was even important
for the study of hadith transmitters, whose varied and distant birthplaces
had to be known. Above all, from a literary viewpoint, it could offer a
treasure-trove of exotic curiosities.

The subject could be studied from two viewpoints. Following the old
Greek and Sanskrit traditions, the globe was marked off in degrees of latitude
and longitude (the latter started sometimes from the supposed Western
Isles in the Atlantic, sometimes from Ujjain, site of a major observatory
in central India). Seven 'climes' were charted according to latitude (and

456 THE CIVILIZATION OF THE HIGH CALIPHATE

hence, in principle, climate) between the equator and the north pole —
answering to our three 'zones'. The exact position of various cities was
determined, as nearly as possible, astronomically. This school of geography
(following the Greeks and still afflicted with their incapacity to measure
longitude well, before the day of good watches) exaggerated the length of
the Mediterranean and made it the equivalent of the Indian Ocean, which
was closed in with an African coastline that kept on east, instead of turning
south, from the Horn of Somalia (despite the occasional protests of seamen,
which scholars barely acknowledged). Even so the geographers added
greatly in detail to what the Greeks had been able to accomplish.

Despite failures of calculation, educated Muslims had a reasonably clear
idea of the substantive dimensions of the Afro-Eurasian historical complex.
This was expressed in the second viewpoint from which geography could
be studied, one less congenial to the mathematical bent of the Faylasufs
but more practical for literary purposes and perhaps more realistic. Following
the lead of a man of Balkh who wrote on geography some years earlier
(evidently largely from literary sources), al-Istakhri (d. 951) of Fars put
together a comprehensive description of the world (especially of the Muslim
lands) based largely on his own extensive travels, as well as on studies of
individual lands that were also being made. Cities were usefully located in
terms of travel distances from town to town ; and the lore of exotic things a
tourist might expect to see — and much that he would only hear tell of, at
best — added colour. His work was widely imitated and supplemented. Maps
were crude, and not standardized by printing. The sensible custom grew of
schematizing drastically the overall maps: e.g., the Mediterranean might
be shown as an ellipse, or even a circle.

Following an old Iranian tradition, with modifications, these geographers
divided the inhabited world (i.e., the Afro-Eurasian landmass) into a number
of great realms, conventionally seven (answering to the 'continents' — three,
in the Old World— still used as arbitrary units in Western geography.)
To each of these realms they assigned distinctive traits. In the centre, to be
sure, were the realm of the Arabs and that of the Persians, thought of as the
most active of peoples and the earliest to embrace Islam. These were flanked
by the realms of two great civilizations, those of the Europeans (Rum,
i.e., Romans, Byzantines) and of the Indians, both noted as ancient homes of
philosophy. (The 'Franks' of western Europe were most commonly not
separated off sharply from the Byzantines, whose cultural lead they long
followed, just as no distinction was made within the vast Indie realm.)
Also generally included in the varying lists of seven realms — or peoples —
were the Turks of central Eurasia and the Negroes of sub-Saharan Africa,
and finally the Chinese, who were known for their technical skills as artificers
and artists. The Muslims exaggerated their own relative military might,
learning, and general place among mankind, to be sure, but in the following
centuries the reality was to approximate steadily more closely to their sense

THE BLOOM OF ARABIC LITERARY CULTURE 457

of importance. In any case, they possessed a more realistic image of the
world than did any other civilized tradition. 2

Shi'r poetry as the consummate skill

If literature was the crowning art for the Arabs, within literature it was
poetry, or shi'r. This was a major legacy of the old- Arabian cultural tradi-
tion, which had imprinted its norms on the Mudari Arabic as literary
vehicle. Whatever the position of Pahlavi verse may have been in Sa.sa.nian
culture, the central place of Arabic poetry in Islamicate culture helped to
confirm the breach between the two.

Poetry was chanted or sung — to conventional tunes — in public; not
(in principle) read in the study. We must keep this in mind when evaluating
its impact. It was not a personal message from writer to reader, couching
in bursting words what a direct prose statement would have been insufficient
to express. It was a graceful and beautiful expression of sentiments that could
be common to a gathered audience expecting an evening of tasteful enter-
tainment. Hence the form must be exactly held to, so that the attention
might be drawn to the detail of presentation without being distracted by an
unfamiliar overall approach. And likewise the substance must be familiar
enough to allow each listener to concentrate on noting how well the thought
had been put, without being distracted by considering overmuch the impli-
cations of the thought itself. This was true even of didactic poetry, in which
wise advice was given: the object was not to present a new idea, but
an old and tried idea in a striking and memorable way.

Poetry is, in effect, intensive verbal composition within formal limits.
We assign the term 'poetry' to whatever type of composition, within a given
language-tradition, calls for the greatest intensity and formality. Such
intensive composition is then appropriate for symbolically evocative state-
ments such as we call 'poetic' in a looser sense. But this is partly because the
listener's expectations have been aroused by the very form. Hence some
continuity of form is a part of the communication itself: it is imposed not
by unthinking custom but by the ways of human perceptivity. In modern
English we have become accustomed to a considerable relaxation of the
formal requirements of rhyme, of meter, and of appropriate subject matter.
But some subtle requirements do stand at least as to rhythm and subject
matter. In Arabic (as generally before modern times) the standards were
more rigorous in most respects. In classical Arabic, what answered to our
word 'poetry' was certainly shi'r; the formal limits of shi'r included not only
rhyme and a fixed number of rhythmic patterns, but also certain set types of
subject matter and of verbal usage.

2 The Islamicate notions of geography, inadequate as they were, were less misleading
than the Occidental notions of the time, which still plague modern scholarship. See
the section on usage in world-historical studies in the Introduction.

458 THE CIVILIZATION OF THE HIGH CALIPHATE

Society accepted a number of conventional types of rhythmic pattern,
meters, which every cultivated man and woman knew the flow of. Some of
these were fairly rudimentary, others reasonably complex, like an interesting
tune. (The meter was based on variation in length of syllables, as in Latin,
rather than on stress, as in English.) The poet was expected to fit words into
one of these meters with sufficient closeness to allow the basic pattern to
emerge. The meter was not necessarily readily evident from only one or
two lines of verse. (Among later Arabs, at any rate, the ordinary person
found it a help to know which meter was being adopted, so as more quickly
to know how to read the words to bring out their flow ; hence the meter was
often mentioned in citing verses.) What was important was that the words
sing well when felt within a given meter. A good poet might be quite daring
in varying the number of syllables or the secondary rhythm elements in
such a way that the feeling of the meter was maintained. (This fact adds
to the difficulty many Moderns find in appreciating the Arabic poetry
as compared, for instance, with the — later — Persian poetry, of a firmer
rhythm.)

Lesser lights needed rules to tell them how far they could deviate from
set patterns without getting lost — and in 'Abbasi times such rules were
worked out in tiresome detail. They may give an impression of artificiality
to what was in origin a realm of liberty. Certain poets were credited with
inventing whole new meters, but this was not normally expected ; if one was
to introduce a new meter, the audience must be, as it were, specially trained
to hear the poet's new work, a task disproportionate to any likely benefits.
For the most part, poets stuck to a very few familiar meters even among those
which were recognized.

Poetry, shi'r, was the one great art adopted from pre-Islamic Arabia
(where it was almost the only major fine art). In pre-Islamic Arabia
the genres for which shi'r was used were relatively few — chiefly the poem of
praise or of boasting, exalting one's own tribe or oneself; the insult-poem,
recited against an opposing tribe; or the lament for the dead, commonly
produced by women. These genres were kept consciously separate: each
genre was carried as a distinct tradition, having had its own moments of
invention and improvement which any subsequent poet might take advan-
tage of, once the public had learned to receive them. Though what genres are
ultimately possible is limited only by the limits of human sensibility and of
the media available, the Arabs, like other pre-Moderns, were so aware that a
given public could assimilate only a restricted range of forms that they even
tried to name the inventors of some genres, and even a very creative man
was not ashamed to work in one of the forms so established.

The most respected genre in ancient Bedouin poetry was the qasidah ode,
a long poem with a fixed series of subjects. It began with regrets over a past
love affair and the traces of the beloved's encampment in the desert; con-
tinued with a ride through trackless wastes, in which typically the things of

THE BLOOM OF ARABIC LITERARY CULTURE 459

the desert, as well as the poet's alleged mount, were described by similes;
then, only, the main burden would emerge, often sheer self-praise. The
qasidah might end with praise of some patron from whom the poet expected
a reward for his efforts. But many poetic pieces were not qasidahs but less
formally developed 'fragments', which went more directly to the point.
In any case, a limited range of subject matter had called for a limited range
of emotions and a limited vocabulary. Within these limits the effect was
perhaps as remarkable as any combination of the given elements could
produce. When a fine pattern was found it was held to.

The poetry had been transmitted orally by reciters (often themselves
poets), who chanted it publicly as it came from a poet's lips and passed it
from generation to generation. In 'Abbasi times what had survived of the
work of the great poets was set in writing by various reciters or even
philologians and became something of a poetic canon; it was then that it
was appreciated not as living commentary on current events or on the
virtues and failings of the tribes, but as a model literature to be pondered
for its subtlety and its richness as a body.

Critics acknowledged the pagan Imru'-al-qays as the greatest of Arab
poets. In a hadith report, Muhammad himself was made to confirm this
judgment — Imru'-al-qays was the greatest of the poets and their leader to
Hell. A dissolute and venturesome grandson of the head of a great central
Arabian tribal agglomeration controlled by the tribe Kindah at the end of
the fifth century, the story of his life stirred the imagination : his father had
inherited an important tribal chieftainship, after the Kindah imperium had
broken up, and had been murdered; and Imru'-al-qays turned from his
scandals with the maidens of Arabia to exact on a grand scale the required
vengeance for his father. Tradition made him a mighty but reckless and
unfortunate fighter, who was finally destroyed by the mistrusting Roman
emperor himself, of whom he had asked aid — he was sent a cloak of honour,
which proved venomously poisoned. His language was pure enough for a
grammarian, but, perhaps at least as important, his life was romantic
enough to carry a bureau-clerk's dreams far beyond the prosaic streets of
Baghdad.

His diwdn (collected verse) was gathered and, so far as possible, verified
by the court grammarian al-Asma'i (d. 831). Imru'-al-qays was (incorrectly)
regarded as having invented the erotic prelude with which every regular
qasidah ode ought to begin; certainly he was especially admired for his erotic
passages — of a frank directness which was at the same time fresh and
unstrained in its images. Critics praised him as the master of simile and
metaphor. The images of simile, indeed, became the treasure of Arabic
poetry, what above all a poet hoped to excel in. His most famous poem was
a qasidah ode included in a famous collection of seven Mu'allaqat, pre-
Islamic masterpieces claimed by legend to have been poems which won
the prize in annual contests at a fair near Mecca; but edited in 'Abbasi

460 THE CIVILIZATION OF THE HIGH CALIPHATE

times. Imru'-al-qays' poem powerfully paints us his passion and his
self-will. 3

In the varied urban society of 'Abbasi times this pagan tradition had been
maintained as an ideal of chaste and integral poetic expression; it was the
standard by which more modern poetry was judged — and usually con-
demned. For some, it represented the specifically Arab heritage — and even
persons not of Arab ancestry might like to be 'more Arab than the Arabs'
by cultivating it. But probably a more subtle appeal was an undeniable
romance and fascination in the verse of men seemingly unbound by the
sober second thoughts of proper city folk, and in particular by the horizons
of Islam, and so expressing untrammelled human feelings.

In the circles of the adibs new poetry continued to be composed. But it
was at least as important to use verses — normally part of a memorized stock
and as old as possible — effectively in the midst of conversation or of prose
composition. Already in the Bedouin poetry, each line of verse had tended
to be a self-sufficient unit expressing a quite generalizable sentiment. Now
such units were used to adorn a letter, a document,; or even a treatise (and
frequently to impress the reader with one's erudition). Yd 'idu, md la-ka
min shawqin wa-irdqi, wa marri tayfin 'aid ahwdli tarrdqi — 'You hardened
[man], who is like you for longing and sleeplessness, and a spectre's coming
amidst the terrors of a far- wanderer', the poet addresses himself. (I simply
take the first line of a famous collection made ■ for al-Mansfir, the
Mufaddaliyydt.) The words in Arabic are concise and direct, but archaic
(by 'Abbasi times) and of a strange turn. They were attributed to a sixth-
century Bedouin outlaw, an uninhibited brigand whose usual name,
Ta'abbata Sharran, means 'mischief under his armpits' (referring to his
sword, carried there). There is no piety in them, but a frank expression at
once of human toughness and of human need and loneliness. At the right
spot, in some clerk's prose, such a line could bring echoes of another world,
and (if the prose were strong enough to carry it) could lend it magic.

For these varied reasons, religious, literary, and social, the qasidah form
itself (and the manner of poetic line derivable from it) took on an air of
unassailable authority as embodying the norms of Arabic language and adab
culture. In doing so, it gave rise to a varied and influential school of literary
criticism, dedicated to guarding the standards thus achieved.

Criticism: the old-Arabian classical norm and the new poetry

The ascendancy of the Arabic poetic tradition thus gave further opportunity
to the Shari'ah-minded opponents of the absolutist tradition to inject some-

3 A. J. Arberry, The Seven Odes (London, 1957), discusses the history of the Mu'allaqat
in detail and also their various translations into English. Before him, the most respected
translator was Charles Lyall ; Arberry 's own translation is very good, but still, like the
other translators, he cannot bring himself to restrict his translation to the concise
directness of the original.

THE BLOOM OF ARABIC LITERARY CULTURE 461

thing of their viewpoint into the heart of adab culture — and, incidentally,
to confirm the breach with the Sa.sa.nian past. The ancient Arabian poetry
was necessarily regarded by the grammarians as the only contemporary and
therefore fully reliable source for linguistic parallels to Qur'an and hadith,
on the basis of which the meaning of words and phrases used in these latter
could be understood. Thus, paradoxically, the 'ulama' scholars gave a
peculiar blessing to the study of the pre-Islamic poetry, pagan though it
might be and very typical of a boastful, Active, heedless luxury which
Muhammad himself had condemned and which many 'ulama' did not find
particularly commendable as practiced in their own time.

To be sure, the 'Abbasi appreciation of the earlier poetry was not much
influenced by such theological considerations; it extended, but only with
somewhat less approval, to the (often rather impious) Muslim poets of the
Marwani period who had continued the old tradition but could not be
regarded as sure to be lexically impeccable. What mattered to the adib was
the exemplification of elegant grammar — and the echo of an olden time of
liberty. But the 'ulama' did matter, and their implicit attitude was surely
felt in matters of propriety in verse.

The debate in poetry was part of a wider debate on the cultural relevance
of the old- Arabian tradition. Many in the courtly circles objected to even the
limited role which the old-Arabian tradition had come to play in the culture
of the empire. They resisted the tendency of the Arabic philologians (and,
behind them, the 'ulama') to undermine the prestige of the Iranian-based
adab tradition of the katib clerks. They pointed out that a large old-Arabian
role could not be justified in the name of Islam — which had been directed
against just the pagan Arabian tradition which now was so greatly honoured.
A whole literature arose praising the merits of other peoples at the expense
of the Arabs and their' tribal pride; the Arabian tradition was condemned
(in Arabic) as uncivilized and lacking in good taste.

This literary movement, called the 'Shu'ubiyyah', the championing of the
(non-Arab) peoples, had a vogue so long as Arab pride still played a role in
the caliphal state. The conscientious researches of philologians like Abu-
'Ubaydah (of Persian Jewish origin — d. c. 825), who did more than anyone
else to standardize the tradition about old-Arabian affairs, were made use
of to show how petty and uncouth the tribal Arabs had often been. But such
an attitude was hotly attacked by those who mistrusted the aristocratic
narrowness of the adab of the clerkly tradition. In particular, Ibn-Qutaybah
of Marv in Khurasan (d. 889), living mostly at Baghdad, showed that a
prejudice against things Arabian was unjustifiable. Profoundly hadith-
minded in his religious views, in adab he integrated old-Arabian and Iranian-
derived materials into a single adab corpus. He studied all the topics of adab,
including hadith (from a philological viewpoint). Of most general literary
interest was a collection of apt selections from hadith and verse and historical
report arranged to illustrate various sorts of topic — many later literary

462 THE CIVILIZATION OF THE HIGH CALIPHATE

guides for the adib were based on it as a model. Soon the chief arbiters 4 of
literary taste decided that Arabian poetry had been the supreme poetry;
that Arabian taste, if not to be imitated, was at least to be respected; and
even that Arab descent was an honour, as indicating kinship with the
Prophet. In effect, it was recognized that society could not be restored to
its cultural patterns as of any given Sasanian century; that the special
culture of the Marwani-age Arabs had become an element in the heritage of
all the Muslims.

Meanwhile in the field of poetry, already by the time of al-Rashid, the old-
Arabian norm was being challenged at its heart. At the hands of such as
Abii-Nuwas, the significant poetry composed and enjoyed (except, doubtless,
in the desert itself still!) was more luxuriant than the older norm. Though a
restricted number of forms were still used — normally every poem must be
rhymed, with a single rhyme being repeated throughout, for instance — poems
might deal with every kind of sentiment so long as it did not become too
biographical but remained generalized. Thus there might be poems of hunting
or of love or of grief or of philosophic resignation or of shrewd wisdom or of
drinking. The luxuriance was limited. The various genres tended to remain
distinct, each within a well-defined tradition; and in any case, each sentiment
must be perfectly generalizable. If a poet sang of old age, he could refer to
white hairs and the scorn of young damsels, he could say that friends were
dead or forgetful; but he could not introduce special episodes that might
have happened to him but could not be considered typical of old age. This
would be to forget the audience and abandon the pure impersonality of art
in favour of merely private anecdote. Yet, as compared with desert poetry,
the new city poetry expressed an enormous variety of interests and senti-
ments, and allowed considerable freedom of form and imagery. 5

Since the time of Imru'-al-qays, the tradition that he had helped render
great had thus developed, transplanted to a new habitat, very diversely.
But eventually the earlier norms of the tradition, always necessarily still
taken note of, were reinforced by the arbiters of taste in adab as the only
norms worthy an erudite adib. Whether a new turn in a tradition can be
fully legitimized — in this case, by the literary critics, as a basis for courtly

4 H. A. R. Gibb, 'The Social Significance of the Shu'ubiya', reprinted as chapter four
in his Studies on the Civilization of Islam, ed. Stanford Shaw and Wm. Polk {Boston,
1962), clarifies the sequence of the establishment of the prestige of Arabic traditions
among the clerks, and many other points by the way. Gerard Lecomte, 'Le probleme
d'Abu 'Ubayd', Arabica, 12 (1965), 140-74, shows how grammatical and adab discus-
sions, in the time of Ibn-Qutaybah, got tied in with theology. (Some of his conclusions
are subject to reserve because of his assumption that there was an 'orthodoxy' at the
time; he even suggests someone might have been a 'crypto'-Mu'tazili in an age when
Mu'tazilJs dominated the field of grammar as well as politics.)

5 Gustave von Grunebaum, 'Growth and Structure of Arabic Poetry ad 500-1000',
in The Arab Heritage, ed. A. Faris (Princeton University Press, 1944), brings out
excellently the differences that supervened generation by generation (though it con-
centrates on the pre-Islamic) .

THE BLOOM OF ARABIC LITERARY CULTURE 463

prestige — depends not so much on how close it remains to the past (every
new step in a tradition jars a bit) as on its relevance to the ultimate expec-
tations of its public. The 'new' poetry was indeed well loved, but the only
basis in that milieu for full legitimation of poetry was as much grammatical
and historical continuity as transient current delight. Many 'Abbasi critics
insisted that only the type of language and the range of themes consecrated
by Imru'-al-qays and his peers could be legitimate in genuine shi'r; the test
of skill was to excel within such limits. They condemned the 'new' poets,
like Abu-Nuwas, who had presumed to introduce not only new motifs but
vocabulary alien to the old Bedouin purity which the critics prized; in
effect, they had refused to attempt the old game and had yet proposed to
express themselves in poetic form, calling their results 'shi'r'.

Among the most respected critics was Ibn-Qutaybah, who strongly
appreciated poetic values as part of adab culture generally. His work
on lexicography, rules of prosody, and grammar strengthened the more
conservative critics. He himself defended Hadithi standards in literature, and
criticized adversely so free a writer as al- Jahiz (whom we shall meet shortly) ,
but he also made a point of recognizing that the newer poetry might some-
times be just as good as that by ancient pagans.

After Ibn-Qutaybah, other critics worked out more ample analyses of
the use of simile and the like, which allowed greater room for the 'new'
poets. The result of this contest of critics was, for a time, an active production
of literary criticism alongside the poetry itself. One of the most adventurous
of the critics was himself a poet. Ibn-al-Mu'tazz (d. 908), the son of the
caliph, wrote up the saga of a later caliph's victories in an epic poem such
as was found in the Persian tradition but not elsewhere in Arabic. He
developed an elaborate critique of the various sorts of tropes and literary
turns, favouring inventiveness. Another critic even made use of the Greek
tradition of literary criticism.

After the death of Ibn-Qutaybah, however, a certain systematizing of
critical standards set in, especially among his disciples, the 'school of
Baghdad'. It was at this time that in fiqh men began to speak not of the
Iraqis and the Hijazis, but of the Hanafis and the Malikis and the Shafi'is;
and every earlier legist was somehow placed in reference to one or another
recognized school. It was also at this time that retrospectively the gram-
matical schools of Kufah and of Basrah were distinguished and their positions
contrasted. Literary lines likewise began to be drawn.

Finally the doctrine of the pre-eminence of the older classics prevailed.
So far as concerned poetry in the standard Mudari Arabic, which was, after
all, not spoken, puristic literary standards were perhaps inevitable: an
artificial medium called for artificial norms. That critics should impose some
limits was necessary, given the definition of shi'r poetry in terms of imposed
limitations. With the divorce between the spoken language of passion and the
formal language of composition, they had a good opportunity to exalt a

464 THE CIVILIZATION OF THE HIGH CALIPHATE

congenially narrow interpretation of those limits. Among adibs who so often
put poetry to purposes of decoration or even display, the critics' word was
law. Generations of poets afterwards strove to reproduce the desert qasidah
ode in their more serious work so as to win the critics' acclaim.

Some poets were able to respond with considerable skill to the critics'
demands. Abu-Tammam (d. c. 845) both collected and edited the older
poetry and also produced imitations himself of great merit. But work such
as his, however admirable, could not be duplicated indefinitely. In any case,
it could appear insipid. A living tradition could not simply mark time; it
had to explore whatever openings there might be for working through all
possible variations on its themes, even the grotesque. Hence in the course
of subsequent generations, taste came to favour an ever more elaborate style
both in verse and in prose. Within the forms which had been accepted, the
only recourse for novelty (which was always demanded) was in the direction
of more far-fetched similes, more obscure references to educated erudition,
more subtle connections of fancy.

The peak of such a tendency was reached in the proud poet al-Mutanabbi',
'the would-be prophet' (915-965 — nicknamed so for a youthful episode of
religious propagandizing, in which his enemies said he claimed to be a
prophet among the Bedouin), who travelled whenever he did not meet,
where he was, with sufficient honour for his taste. He himself consciously
exemplified, it is said, something of the independent spirit of the ancient
poets. Though he lived by writing panegyrics, he long preferred, to Baghdad,
the semi-Bedouin court of the Hamdanid Sayf-al-dawlah at Aleppo ; and on
his travels he died rather than belie his valiant verses, when Bedouin attacked
the caravan and he defended himself rather than escape. His verse has been
ranked as the best in Arabic on the ground that his play of words showed
the widest range of ingenuity, his images held the tension between fantasy
and actuality at the tautest possible without falling into absurdity.

After him, indeed, his heirs, bound to push yet further on the path, were
often trapped in artificial straining for effect; and sometimes they appear
simply absurd. In any case, poetry in literary Arabic after the High Caliphal
Period soon became undistinguished. Poets strove to meet the critics' norms,
but one of the critics' demands was naturally for novelty within the proper
forms. But such novelty could be had only on the basis of over-elaboration.
This the critics, disciplined by the high, simple standards of the old poetry,
properly rejected too. Within the received style of shi'r, good further work
was almost ruled out by the effectively high standards of the 'Abbasi critics. 6

6 H. A. R. Gibb's masterly Arabic Literature, an Introduction (first ed., London,
1926; new edition, 1963), lists in its appendix a good many translations of Arabic literary
works into Western languages. Several organizations have latterly been adding to the
list, notably the Unesco 'East-West' project. Among anthologies, I shall mention
Reynold A. Nicholson, Translations of Eastern Poetry and Prose ('Eastern' here means
Arabic and Persian, and the prose is mostly historical) (Cambridge University Press,
1922) — elegant and by a master, but suffers from a preoccupation with making the

THE BLOOM OF ARABIC LITERARY CULTURE 465

Prose and saj' : entertainment and edification

Arabic prose, which had begun as a direct inheritance from the Pahlavi of
the Sasanian absolutism, did not maintain this tone. Possibly it played a role
comparable to what prose may have played socially at the Sasanian court,
but certainly it did relatively little to focus the ideals of the adib on the
great monarch — whereas almost all the secular literature we have from the
Pahlavi tends to centre on the monarchy. The interests of the adib were
eclectic and cosmopolitan. Prose was no more royalist than poetry.

Like verse, prose also, of course, is formal and relatively intensive verbal
composition. It is not simply 'conversation written down'; its appropriate
forms must be cultivated. Prose too has its genres and conventions. But we
use the word for less formal and less intensive compositions than the type we
call 'poetry', for it answers to the need for more flexible means of expression.

As we have noticed with regard to poetry, the purpose of literature
generally was not, in the Agrarian Age, to enable a tormented spirit to pour
forth its soul; nor was it even to set forth every sort of information which
might arouse someone's curiosity. Even in our time there are some limits
imposed not only by public censors but by a general, if tacit, sense of
literary relevance. Such limits were always stricter in the Agrarian Age,
though they varied in detail from society to society. Some types of literary
expression were cultivated in ancient Greece, for instance, and not in
Islamdom; notably the tragedy, which allowed for an expression of cosmic
fate in poignantly human terms, but which Christianity had no place for and
which Muslims had no cause to revive. (Islamicate literature, indeed, con-
sistently avoided the personally poignant.) Other types of literature found
in Islamdom were wanting in ancient Greece. But always only a relatively
small range of themes was acknowledged as appropriate for literature ; in
particular, the more intimately personal themes, which have modernly
become so fundamental among us in novel and autobiography, tended to be
excluded, for the most part, as not fit for public display.

Verbal cultivation in Abbasi times among the adibs had two major
functions. On the one hand, it was highly refined play; making a poem or
telling a tale was judged on rather the same basis as playing a good game
of chess; and both were perhaps taken more seriously as the occupations of
a gentleman than they are in a modern business community. The elegance

renderings sound superficially like what nineteenth-century Britishers would recognize
as poetry (a very common sort of difficulty); James Kritzek, Anthology of Islamic
Literature (New York, 1963) — in majority Arabic, with much Persian and some Turkish,
including prose as well as verse, with the advantage of variety not only in substance
but in translators; Herbert Howarth and Ibrahim Shukrallah, Images from the Arab
World (London 1944) — an impressionistic set of freely rendered fragments (including
some from non-Arabs who happened to use Arabic), and especially A. J. Arberry,
Arabic Poetry, a Primer for Students (Cambridge University Press, 1965) — literal but
intelligible renderings facing the Arabic text, with a helpful preface on prosody and
literary figures. There are several useful anthologies jn French.

466 THE CIVILIZATION OF THE HIGH CALIPHATE

of one's home appointments and the elegance of one's epistles — especially if
one were a man of position — went far toward setting off one's rank, or rather
one's worthiness of that rank. On the other hand, literature had the function
of edifying. It should give expressly useful information, either rules for a
person's behaviour or facts of an exemplary sort which he might have
occasion to learn from for the future. Even sheer curiosities were presented
in the guise of material which will increase one's urbanity or lend colour to
one's epistles. Despite the high rank allowed to poetry, both these functions
were fulfilled for the most part in prose.

The adib should, among other things, have a good command of proverbs,
preferably those originating with the Bedouin ; and know the favourite tales
of fiction — such as those in the Thousand and One Nights collection, already
available in primitive form then, and the tales of the absurd Juha, a clownish
figure through whose antics sometimes a simple wisdom can shine. (Such
tales were not so highly respected as the proverbs, however.) Ibn-al-
Muqaffa"s translation from the Pahlavi of a book of moralizing stories
(animal fables), under al-Mansur, became a standard of style for subsequent
writers to imitate.

This prose was simple and direct and admirably calculated to retain the
attention with a clever interplay of image and of idea. The tradition of
simple, clear prose found its most prominent exponent in 'Amr al-Jahiz
(c. 767-868), who made conscious use of variety in pace and in level of
thought in order to prevent boredom, as he said, and to make his work
serve both for amusement and for edification. But whereas Ibn-al-Muqaffa"s
work was at least set at the lion-king's royal court, al-Jahiz' work was
relatively unconcerned with princes. By combining informativeness and
literary grace, however, he met the social needs of the adib perfectly.

Al-Jahiz was ugly: his eyeballs protruded (whence his name) and Arab
society held against him his Negro ancestry. He was also bitter of temper
and evidently did not fit easily into official jobs. His gift for the apt phrase
and the telling anecdote, and the comprehensive information he could draw
on to back up any point, made him inordinately famous even in his own day.
But it is said that when the caliph al-Mutawakkil sent for him to be tutor
to a son (a post usually reserved for the chief scholars of the time), al-Jahiz'
looks so appalled him that he sent him away immediately with a handsome
gift in recompense.

Al-Jahiz delighted in anecdotes ; his Book of Misers lists every breed of
that unpleasant but variously eccentric species, illustrating his points with
appropriate tales, some hung on prominent personalities and all allegedly
true. The genre presumably continued in Arabic a tradition going back to
the Greek Characters of Theophrastos. Aesthetic distance, allowing the reader
to appreciate the absurdities without having merely to relive them, was
maintained by selecting only those traits in the personality that expressed
or pointed up the 'character' of a miser as such. Yet al-Jahiz' descriptions

THE BLOOM OF ARABIC LITERARY CULTURE 467

are endlessly and subtly individualized. He can tell of a legendarily extreme
miser, a wealthy man who explains that he cannot in conscience exchange
coins, on which is inscribed the shahadah declaration of Islam, for mere
unhallowed goods without even the name of God upon them (this tale is
climaxed when, on the miser's death, the son who receives the inheritance
examines everything and then — discovers a way to economize still further
and condemns his father for a spendthrift). But more appealing is the
picture of a reasonably decent man (a personal friend of the author's,
deceased) who just cannot bear to spend money on hospitality for his
friends — serving one's guests large amounts to eat, of course, was the heart
of Agrarian Age hospitality, when wealth meant food — and is wonderfully
inventive in keeping them from eating much; and even manages to justify
his ways as representing the truest hospitality.

Al-Jahiz wrote innumerable tracts attacking or defending almost every
party or group with memorable effect. He thus wrote on the virtues of Turks
and of Negroes but also defended the superiority of Arabs over other
peoples; he defended the reputation of Jews as against the then more
honoured Christians; he attacked Mazdeism, maintained the superiority of
merchants over officials, and compared the excellences of lads with those of
maidens. Most of these little essays were designed primarily for amusement
{though it is claimed he sometimes wielded his pen against someone for
money) ; yet al-Jahiz had serious claims. His curiosity led him to make little
experiments to disprove various popular superstitions, his comments on the
Qur'an were likewise on the side of reasonableness, and he could undertake
a balanced and even convincing defence of an unpopular position which he
did not himself hold — thus his presentation of the Shi'ah picks out not the
most absurd traits (as was common with Jama'i-Sunnis) but the most
defensible positions that a Shi'i might maintain, and outlines them with
urbanity. He was an eager Mu'tazili and is credited with sufficient theological
acumen and influence to have established a special Mu'tazili school of
thought.

His longest surviving works, however, are serious in a different way. They
display unbounded philological and literary erudition, providing a store-
house of the sort of information that a katib clerk would need so as to write
elegantly, knowledgeably, and exactly. Even the telling of anecdotes seems
subordinated to illustrating technical points of good style. But it was
subordinated more specifically to an Arabic good style, replete with old-
Arabian references. Jahiz' works were probably instrumental in winning
the clerks to accepting the Arabian tradition as a prime resource of adab,
and this may have been his conscious intention. It is such erudite works, at
any rate, that were regarded as peculiarly valuable by the adib. In contrast
to al-Jahiz' lighter prose, such works could be far from simple to read. 7

7 Charles Pellat, Le milieu basrien el la formation de Djdfyiz (Paris, 1953), is an in-
valuable study of social, religious, and intellectual life. It is marred by projecting back

468 THE CIVILIZATION OF THE HIGH CALIPHATE

More honoured than the tradition of Ibn-al-Muqaffa' and al-Jahiz even-
tually, however, was a different prose tradition, that of saj', rhymed prose.
Here the sentences were, if possible, cadenced; in any case, every two or
sometimes three phrases or clauses were made to rhyme. Recherche words
were preferred. The Qur'an already presented a model of such rhyming, but
the literary effect of saj' was more formal and artificial than that of the
Qur'an. But it could be pleasing, and was used for sermons and state letters
alike. Ibn-Durayd (d. 933) used rhymed prose for a compilation of anecdotes
and reports characterizing Bedouin ways and Bedouin events — combining
successfully a fashionable subject matter with a fashionable style.

Saj' prose was marked by a degree of formalized pattern and of care for
verbal expressions and their sound that might in another culture have been
included in the realm of poetry. In fact, however, shi'r, poetry, was in Arabic
so strictly delimited as to permissible forms that there was no tendency to
intergrade between shi'r and saj'. Rather, saj' remained associated with
ordinary prose and very gradually came to influence even the writing of
history and of private letters. It reached its fullest flower, however, only
after the fall of the caliphal state.

The line between saj' and simpler prose was not sharp, in any case. Even
al-Jahiz could use rhyme in his prose, and the delight in balancing one word
against another at once in sound and in sense was endemic. The structure
of the Arabic language lent itself to this. As in other Semitic languages,
compound words were formed from simpler ones not so much by prefixes
and endings as by internal transformations in the sound. Each word could
be analyzed into a root of (usually) three consonants, which could appear
in a large number of set patterns, varying as to vowels and supplementary
consonants ; each of which tended to carry semantic implications : thus from
S-L-M could be formed muSLiM, an active participle, and iSLaM, a verbal
noun — and many other related words and word forms, such as plurals and
pasts. These patterns were strikingly consistent in their formation: thus
muLHiD (a heretic) is likewise an active participle, and iLH&D (the act of
being a heretic) is a verbal noun. Hence both rhythm and rhyme tended to
point to syntactical and even semantic meaning. Such words can echo each
other down a page in far more intricate resonances than any mere cor-
respondence of endings can produce. This is a feature of Qur'anic style that
helps make it hard to translate, and was almost inescapable throughout
Arabic writing. Parallelism of sense very readily produced parallelism of
sound.

The identification of sound and sense which the naive speaker of a single
language always does feel was thus reinforced in Arabic. Even the
philosophically-minded were sometimes tempted to see in Arabic words

too early a later 'Sunnism' which he posits as immemorially orthodox, and even by an
error too frequently found, of assuming that a 'heretic' must also be a libertine, as
accused.

THE BLOOM OF ARABIC LITERARY CULTURE 469

more than arbitrary conventional signs; and the tendency of every society
to objectify its symbols — from the signs of the zodiac and the hours of the
day to the letters of the alphabet — was intensified among Muslims, who
often saw Arabic as embodying a natural order even when they knew other
languages than Arabic as well. This had occasional consequences in theology
and philosophy which we shall take note of. But its most immediate and
persistent consequences were in the field of literature itself. The gram-
marians elaborated these traits into a system, the many exceptions to which
must be accounted for (in this way they incidentally made Arabic grammar
seem far more difficult than it actually was). They left the impression that
the language was a closed whole: only a limited range of sound patterns
could be admitted in genuine Arabic — and even the lexicon itself seemed
derivable by a set of grammatical rules, once the basic sense of each three-
letter root was given. The high regard in which the language of the Qur'an
was held reinforced this impression. Thus the immutability of classical
Mudari Arabic, and with it of the ancient poetic forms sanctioned by the
critics (already intimated when that classical Arabic was retained for
literary purposes although it was not the actual spoken language) was
further confirmed. Even without this intervention of the grammarians, the
nature of Arabic gave an irresistibly distinctive flavour to Arabic writing,
which encouraged the manipulation of words for the sheer delight of it (and
sometimes misled writers into verbal temptations) . The closed system of the
grammarians then added to this tendency a rigid verbalistic classicism
which eventually discouraged the literary use of Arabic by any who had an
alternative.

The arts of luxury

The verbal arts were regarded most highly, but an elegant decor required
full exploitation of the visual arts too. At the caliphal court and also at some
governors' courts, enormous resources were available for artistic creations,
notably monumental building. Yet the visual arts, like the verbal, were more
cosmopolitan than monumental. In contrast to the Sa.sa.nian tradition,
which had maintained an elaborate continuity of royal iconography not only
in its great buildings and in its stone carving but even in its more luxurious
silverware or textiles, under Islam the caliphal office was almost never the
focus of artistic symbolism.

The artist was an artisan, working to the order of wealthy patrons. In the
normal manner of agrarianate-level society, both the stock of patterns and
the technical skill were handed down from generation to generation within
specialized families. Selection of the craftsmen was not chiefly by artistic
temperament but by birth. Learning the techniques, the firing or glazing
of the pottery, the weaving of the cloth, etc., and learning the particular
shapes and designs to be used formed a single process in training the young.

470 THE CIVILIZATION OF THE HIGH CALIPHATE

Hence there was little occasion for drastic individual innovation from an
artistic point of view; each town had its particular hereditary designs for
which it was known. There might seem little artistic reason for change in
style or substance.

Nevertheless, the existence of a major state like the caliphate had the
effect of bringing out and developing an artistically distinctive style-complex.
Like any other tradition, that of artistic standards and fashions always did
change constantly, if gradually. The maintenance or renewal of high
standards in each new generation depended on the taste and the wealth of
the leading families, who were the patrons. The prosperous court, as well as
the wealthier private persons, such as merchants, could afford to pay for
technical refinement in what they had about them, refinement that cost
endless time on the part of the craftsmen. At the same time, their taste, to
which the craftsmen must cater, was susceptible of developing with the
fashion. The fashion might call for religious or heraldic symbolisms, or might
avoid them. But whatever other interests art served, in any agrarianate-level
society it always filled the need for giving a rich decor to wealthy establish-
ments, public or private. If it was not monumental in the Sasanian tradition,
it could still find its own ways to a renovation of taste.

The art of the High Caliphate was an aristocratic rather than a bourgeois
art, yet it was evidently inspired by little sense of family tradition but rather
by love for a rich immediate setting for whoever might be able to enjoy it.
This immediate accessibility was accentuated by religious scruples. Painting
in Islamdom — in direct contrast to the case of Medieval Christendom — had
as little as possible to do with formal religion. Inheriting a suspicious fear of
idolatry from earlier Jews and Christians between Nile and Oxus, the
'ulama' scholars, Jama'i-Sunni and Shi'i, barred the representation of
animate figures in any religious connection. (We will go into this iconophobia
and its consequences in more detail later, when we treat of the more fully
developed periods of Islamicate art.) Above all, figural art must not be used
for worship — which use, among most other peoples, has been a prime source
of profound artistic inspiration. Consequently, painting was a worldly luxury
of courtiers and rich merchants. In an age when the deeper searchings of the
human imagination expressed themselves mostly in one or another sort of
religious terms, this put a severe initial limitation on the possibility of any
more deeply interpretive visual art arising than what might be expected from
the irreducible personal elan of the craftsman-artist, and from the social
traditions in which he worked.

In the High Caliphal Period, the graphic arts were often beautifully
sustained, but do not seem to have developed such major creative forms as
became possible later. The wall paintings at Samarra developed yet further
the Late Roman and Sasanian tendencies toward a stylized recasting of the
old Greek naturalism. The colourful dancing girls, for instance, showed
traces of the old Greek handling of drapery in their garments, but their

THE BLOOM OF ARABIC LITERARY CULTURE 471

simplified faces and abstract postures were intended not to invoke an
illusion of life but rather to elicit from the dance those lines which would
most gracefully symbolize it statically on what was frankly a decorated wall.

A most pregnant tendency was displayed in the art of movable objects.
During Parthian and especially Sasanian times, the ceramic arts had been
little patronized by the wealthy, especially east of the Iraq. Even in villages,
the pottery remained undistinguished as compared with that of earlier
centuries. But under the 'Abbasids porcelains imported from China (in its
expansive T'ang period) inspired a revival of ceramic art. The porcelain
could not be duplicated, but ways were found to imitate its whiteness.
Then in ceramics as well as in the other craft arts, metal and wood work
as well as work in cloth, the resources which had been being developed by
designers throughout the region — from Coptic plaster-workers and weavers
in Egypt to silversmiths in Iran — were brought together in a new art with
its own traits.

Freed of many of the symbolic implications which seem to have bound
artists in some of the earlier traditions, Islamicate artists — of all confessions, '
though working largely for Muslims — created a new decorative tradition.
The motifs of the earlier arts with the greatest geometrical potentials were
selected for development: for instance, the old motif of a 'tree of life', which
was simplified into a doubled curving pattern which then lost all tree-like
character and — with new branches growing out of its tips — unfolded into a
repetitive pattern indefinitely reproducible in all directions. Motifs popular
in the Mediterranean had long lost much of their illusionistic naturalism.
Now what was left of that naturalism was scuttled, and with it the habit
of confining all form within borders and medallions. An 'all-over' art was
cultivated of potentially endlessly symmetrical interweaving patterns, which
had the virtue of giving an integrally 'rich' effect.

In this art, varied as it was, the Arabic calligraphy introduced a thread of
unity. Calligraphy itself was considered a major art — great calligraphers,
who introduced new styles of forming the script, were more famous than
great painters. 8 Calligraphy formed a unifying theme among the various
sorts of art media, and even within the individual work itself. The Arabic
writing, most of whose characters were made by one or two lines at most,
often repeated, and some of them long or sweeping strokes, could readily
be exaggerated or stylized without losing in intelligibility. Many of its
characters were linked to each other, in forming words, in a smooth flow,
which added to the sense of continuity of design which parallelism of form
could already produce. A band of Arabic writing marching across a door or

8 The conventional Western distinction between 'major' and 'minor' arts — sometimes
still encountered in discussions of Islamicate art — has, of course, no relevance in other
cultures: in each setting, some sorts of media may acquire special prestige and play a
major overall role. The peculiar expressiveness of a creatively beautiful hand is hard
for those to appreciate whose culture regards such accomplishments as mere detail.

472 THE CIVILIZATION OF THE HIGH CALIPHATE

circling around a platter could thus replace a border in creating the impression
of an integrated unit but without putting an end to its sense of indefinite
extendability. The splendour of the court was reflected in the richness of
the art that surrounded it.

In the new capital of Samarra, built at the direction of al-Ma'mun's suc-
cessor, some traces of the monumental architecture of the age have survived. 9
The scale built on was tremendous — the pillared area of the great mosque
was larger than that of St. Peter's at Rome (though it did not, to be sure,
support a dome). The artistic style was strong and simple, with a sense of
decorative line growing out of Aramaic and Sasanian work. But building
seems to have been almost uniformly hasty ; the wealthy were building rather
for themselves than for posterity, and demanded quick results. Brick —
which can be manipulated fairly readily — was the universal medium, even
in Egypt, where (unlike the Iraq) stone is all around. Decorations were often
in plaster. The mighty, enduring stone constructions of Sasanian times were
not reproduced, and most building of the time has long since passed away. 10

' Some writers speak of Islamicate architecture as becoming more Iranian under the
'Abbasids; but this is to confuse architecture at the capital with the architecture of
Islamdom generally. In Syria, the style remained the same under the 'Abbasids as
under the Marwanids; and we have no evidence on Marwani-time architecture in the
Iraq.

! ° K. A. C. Cresswell's masterly and painstaking analyses of the early Islamicate
arts of building (e.g., Early Muslim Architecture, 2 vols. [Oxford, 1932-40]) have enabled
us to understand in some detail what went on in an art only traces of which are left.
Unfortunately, for all his exact measurements and sorting out of 'influences', he seems
to have no sense of art as such, and in this respect we are still, accordingly, largely at a
loss.

The Dissipation of the Absolutist
Tradition, 813-94^

After generations of prosperity and with the integration of the city masses
into the Muslim community, the old Arab impulses which had formed the
empire had become little more than vestiges, visible chiefly in the pedigrees
of some families and in the classical cast of Arabic poetry. The empire had
to be perpetuated on a more permanent basis. Al-Ma'mun (sole ruler,
813-833) and his successors were ruling at the height of the development
of the Shari'ah law: al-Shafi'i, the legist, died in 820; al-Bukhari, the hadith-
collector, died in 870. They were also ruling at the peak of the formation of
Arabic literary culture. Both 'ulama' and adibs, in their own ways, supported
the caliphal state or at least the position of the caliph. Yet the 'ulama.'
legists were doing, in practice as much as possible to reduce its authority
to a tolerable minimum ; while the cosmopolitan literary culture of the adibs,
avoiding the monumental, and being reasonably pious, did little that would
serve to legitimize a land-rooted agrarian absolutism as such. By the time
of al-Ma'mun the court had, in fact, worked out the broad lines of an
absolutism which seemed to satisfy the taste of the adibs. But this was at
the expense of risking ultimate alienation from the 'ulama.', whose standards
the adibs themselves acknowledged.

In an agrarian society in which legitimacy was conferred by Islam, five
sorts of solution might be offered to the problem of creating an enduring
government. The Kharijis and the Zaydis were proposing one sort of solution :
to reconstruct the early Muslim creative source of the tradition in a face-to-
face community, on the basis of a very personal responsibility of the caliph
to the Muslims at large. The Zaydis allowed for a more institutionalized form
of this than did most Kharijis, but in neither case does the solution seem
to have been adequate to a large-scale international society; neither Zaydis
nor Kharijis showed signs of ability to erect more than local states, and those
in relatively isolated places. Secondly, the Isma'ilis, by way of the batin
interpretation, hoped to derive from the basic monotheistic principles of
the tradition a quite new political system, which might have been more
adaptable than the Medina pattern to large-scale agrarian society. But they
were not able to persuade most Muslims that their hierarchical system of
legitimation was sufficiently true to the original Qur'anic and Muhammadan
inspiration. Thirdly, the Faylasufs proposed, in effect, that Islam should
play the role of a political mythology to support a Philosophically-conceived

473

474 THE CIVILIZATION OF THE HIGH CALIPHATE

state; but such a solution would actually have meant a reversal of the
decisive position gained by the prophetic tradition with the victory of Islam,
and never attracted many adherents. By far the most popular solutions
were those of the adibs, in their pro-absolutism, and of the Shar'i 'ulama',
whether Jama'i-Sunni or Twelver Shi'i (who rejected the activist policies of
other Shi'is).

The adibs had generally no explicit policy for solving the relations between
Islam and the state, but the tendency of their implicit policy is clear: left
to themselves, they would have adapted Islamic terminology to old agrarian
monarchical principles (thus they were willing to call the caliph 'Shadow of
God' on earth, almost implying a mediatory role for him), and might have
subordinated the religious specialists to the authority of the court and its
bureaucracy. Whatever their prospects of success in this, however, they
either would not or could not go so far as to erect an authoritarian religious
hierarchy on the Mazdean model of Sa.sa.nian times. The 'ulama', on the
contrary, lauded any resistance that was offered to such tendencies — for
instance, the staunch refusal of some 'ulama' to address the caliph in any
more obsequious way than as a fellow Muslim ; they developed those aspects
of the overall Islamic tradition that depended on initiative in society at
large, as against those that depended on political leadership, and would
have reduced the caliph to an administrator of the Shari'ah, first among
essentially equal Muslims. In fact, of course, the caliph had become a figure
radically alien to such principles, a figure whom they had to hem in but
could not genuinely absorb.

The comprehensive demands that the 'ulama' were perfecting breathed a
spirit almost directly opposed to that of the adibs. For the adibs, social
rank and privilege were of primary importance, however fluid they might
be ; high culture was the end of social organization ; the litterateur with his
inventive gifts was the model of living. For the 'ulama', almost any sort of
special privilege (save possibly that of the family of Muhammad) was ruled
out; justice among ordinary men and women was the end of social organiza-
tion; the hero was the man who conformed most closely to a moderate
pattern of productive common life. Though the standards of the adibs were
alone very effective on the level of day-to-day politics, those of the 'ulama.',
embodied in the Shari'ah, alone received long-run respect, at least in their
chosen sphere, even among the adibs themselves; respect of a sort that
would enable the Shari'ah, even in crises, to maintain its prestige — that is,
to enjoy a general expectation that people at large would remain faithful
to it despite temporary setbacks.

Al-Ma'mun: experiments in aligning the Shari'ah and the court

Al-Ma'mun was both intellectually curious and piously concerned with
justice. Yet he fully satisfied neither adibs nor 'ulama.'. Having begun his

DISSIPATION OF THE ABSOLUTIST TRADITION 475

reign in Khurasan, he ruled through Khurasani officers and, indeed, long
delayed coming to Baghdad at all. But in any case, he was not popular at
Baghdad — despite the traditional patronage of the dynasty, more than one
poet opposed him. Even apart from such explicit discontent, he was intensely
aware of the precariousness of the empire's unity, and concerned to strengthen
it. His reign had started in the midst of the uncertainty and civil war let
loose by al-Rashid's notion of dividing the empire. His victory had undone
the immediate effects of al-Rashid's attempted division, but the unsettling
of minds, that had resulted, persisted and demonstrated the lack of any
political consensus among the population, especially among the Muslims.

In Egypt, the violence which had broken out during the main struggle for
Baghdad itself persisted for years, with rival factions quarrelling in practical
independence of the central power. More important, for the first six years of
al-Ma'mun's reign his control of the Iraq remained insecure. A great Shi'i
rebellion, led by an adventurer, Abu-1-Saraya, at Kufah and Basrah (814-
815), gained the support of Mecca. (It was the only movement that gave
serious indication of stirring an all-Muslim consciousness.) Months after that
was suppressed, the hostility of Baghdad to Khurasani rule reasserted itself.
Al-Ma'mun's long stay in Khurasan led the 'Abbasi leadership at Baghdad
to rebel (816-819) and even to try to set up another caliph (Ibrahim, 817-819)
who would stay in Baghdad; but al-Ma'mun's good generals and his control
of most of the provinces enabled him to quash this step, though he did find
it necessary to move to the recognized capital (819). Meanwhile, in addition
to the Khariji rebellion already going on in Khurasan, popular movements
among the Iranians of Azerbaijan, the Arabs of the Jazirah, and the Copts
of Egypt were expressing a widespread impatience with central control.

Subtler was another danger to the absolutism. After most of the overt
revolts had been suppressed and Baghdad itself fully pacified, in the latter
part of the reign, governors in the Maghrib (the Aghlabids) and even in
Khurasan (the Tahirids), who had been suppressing still other local rebellions,
found themselves able to behave within their provinces like local dynasts —
and indeed founded hereditary dynasties which, however, maintained close
ties with the caliphs. Several families had won such loyalty among the
troops that their members were found to be appropriate as governors in
widely scattered areas — as were, for instance, the Sajid family. When this
independent prestige was combined with a particular territorial base, it
threatened the central power. The general Tahir and his family, in particular,
were forming almost a state within a state. Tahir was military governor of
Baghdad, the submission of which he had assured; but, with local alliances
in Khurasan, he remained at the same time ruler of Khurasan and its vast
dependencies; and his son succeeded him without any question. That is,
the most important troops in the empire, the Khurasanis — more effective
than the still important Arab troops — were controlled by a family that
had its own local basis of power in a single province, a province at once

476

THE CIVILIZATION OF THE HIGH CALIPHATE

The Weakening of the High Caliphate, 813-945

THE GENERATION OF AL-MA'mOn'S TIME

813-833 Caliphate of al-Ma'mun: court favor to new translations of

Falsafah works from Greek; attempt to suppress the new
Ahl-al-Hadith piety (belief in uncreated Qur'an) with support
of Mu'tazili kalam (theological discussion)

780-855 Ahmad b. Hanbal, disciple of al-Shafi'i, persecuted by al-Ma'mun

as symbol of Ahl-al-Hadith resistance to Mu'tazili orthodoxy;
he became imam of an intransigent school of fiqh

816-837 Revolt of Babak against large landlords and Arabs, centered

in Azerbaijan

819-873 Tahirids, as hereditary governors, autonomous in Iran and the

east

833-842 Caliphate of al-Mu'tasim: introduction of Turkish mercenaries

as basis of caliphal power (capital moved to Samarra.)

GENERATION OF THE SAMARRA CALIPHS (836-892)

842-847 Caliphate of al-Wathiq, last great 'Abbasi ruler of a relatively

unweakened empire

869 Death of Ibn-Karram, Muslim evangelist of Khurasan

870 Death of al-Bukhari, author of the greatest collection of
canonical hadith reports; paired with that of Muslim (d. 873)

847-861 Caliphate of al-Mutawakkil, who abandoned the caliphal

attempt to prescribe theological orthodoxy through
Mu'tazilism, and gave support to Ahl-al-Hadith piety;
persecuted the Shi'is; first caliph to be murdered by his
Turkish soldier corps

861-945 Breakup of 'Abbasi power; province after province becomes

essentially independent, till at the end the caliphal government
loses all territorial power; economic and social prosperity persist
or increase in most areas

861-869 Three caliphs in the hands of the Turkish soldiery; all but the

central provinces break away from effective control

861-910 Saffarids (military family) ruling independently in eastern, and

sometimes in most of, Iran

864-928 Zaydl Shi'l state in sub-Caspian lands

868-906 Ibn-Tulun, and sons, practically independent governors in

Egypt
869-883 Revolt of Zanj (African) slaves in the lower Iraq

892 Death of Muhammed al-Tirmidhi, collector of hadiths and

systematizer of categories in isnad criticism

869-892 al-Muwaffaq, executive brother of Caliph al-Mu'tamid,

re-establishes caliphal authority between Syria and Khurasan

DISSIPATION OF THE ABSOLUTIST TRADITION

477

GENERATION OF AL-MUWAFFAQ S RESTORATION

875-998 Samanid (Persian) governors semi-independent, but usually in

accord with caliphal policy, in Transoxania

873-940 The twelfth imam of the Twelver Shi'is having disappeared, he

is represented by four wakils, in the Lesser Ghaybah; after
which, in the Greater Ghaybah, the Twelvers lose contact with
their imam

838-923 Ibn-Jarir al-Tabari, commentator on the Qur'an; and historian

(his history reaches to 913) using chronological sequence; in
contrast to his great contemporary, Ahmad al-Baladhuri
(d. 892)

890-906 Qarmatians, Arab bands of Isma'ili Shi'is, active in 'Iraqi and

Syrian deserts; they mark the beginning of rise to power of
various Shi'i groups, thus hastening the collapse of caliphal
control

892-908 Al-Mu'tadid (to 902), son of al-Muwaffaq; and then al-Muktaft;

caliphal control stabilized from Egypt to western Iran, with
Aghlabid governors in North Africa and Samanid governors in
eastern Iran tributary but not under control

GENERATION OF THE QARMATIAN REVOLTS

894 Qarmatian Shi'i state in east Arabia founded

c. 900 Zaydi state in the Yemen founded

900 Samanids under Isma'il (892-907) add Khurasan to their rule,

upon ousting the Saffarids; soon become patrons of Persian
language works

908-932 Caliphate of al-Muqtadir; caliphal authority again founders on

misrule; Shi'i families in considerable power at capital

905-979 Arab Hamdanids autonomous, and sometimes independent, in

Mosul (and later Aleppo) ; they support Shi'ism

909-972 Isma'ilis establish Fatimid caliphate in Maghrib, replacing

Aghlabids
929-1031 Umayyad rulers in Spain assume the caliphal title (912-961:

caliph 'Abd-al-Rahman III re-establishes Muslim united power

within Spain); Spanish Arab culture flourishes so as to attract

local Christian participation
913-942 Arabic and Persian culture both favoured under Nasr II,

Samanid in Transoxania and Khurasan (940, death of Rudaqi,

first classical Persian poet)

873-935 al-Ash'ari, reconciles the methods of Mu'tazili kalam with the

dogmas of the Ahl-al-Hadith

934-1055 Rulers of Persian Buyid military family independent at

separate centers in western Iran and the Iraq using Persian
(Daylami) and Turkish soldiers; favour Twelver Shi'ism

935-945 Diminishing remnants of caliphal power wielded by an amir

al-umara, military chief, till that office is taken over by Buyids
occupying Baghdad

937-969 Ikhshidid governors independent in Egypt

478 THE CIVILIZATION OF THE HIGH CALIPHATE

remote from central control and of primary importance in the empire as a
whole.

This answered to more fundamental social facts: whereas the Khurasani
troops who put the 'Abbasids in power originally had still felt themselves
largely as Arab settlers, those who supported al-Ma'mun were frankly
Iranians. 1 But Baghdad and the Fertile Crescent were not prepared to
submit to a Tahirid and Khurasani domination — nor was al-Ma'mun him-
self; nor was it clear that Khurasan was willing to submit to domination by
any Baghdad-based force either. Everywhere, relatively local loyalties were
prevailing.

There was still a general sentiment among the Muslims for Muslim unity,
despite the growing importance of relatively local interests. Thus the rival
factions in Egypt had depended for psychological support, as well as in a
more material way, upon the powers at Baghdad, whose active intervention
during the war of succession was readily effective when it came. In the Iraq
itself, naturally, there was an unceasing demand for a single caliph of all
Islamdom; the fact that al-Ma'mun won the allegiance of other provinces
and especially of the holy cities in the Hijaz was a major advantage for
him in his original struggle, with al-Amin. But this sentiment for a united
caliphal state was obviously not fully dependable.

Al-Ma'mun was an able ruler; choosing effective generals, he was able
to bring under control most of the risings that defied him. He maintained a
persistent military struggle for mastery; but he also launched an effort to
tighten the religious base of the empire. This could take the form only of
some understanding with the Shar'i 'ulama' which would not compromise
the power of the absolute monarchy.

Al-Ma'mun was of a serious and inquiring turn of mind, which in his
courtly patronage led him to the increased fostering of natural science and
philosophy that we have noted earlier. On the political level, the same trait
led him to religious policies that might have had the effect of shoring up
the courtly imperial ideal with an officially recognized religious establishment
in which as many Muslims as possible could unite. Whether he ever had in
mind the example of the hierarchical Mazdean high priests, who balanced
the Sa.sa.nian monarchy in social authority and, by and large, gave it effective
independent support and continuity, his efforts, if successful, could even
have led toward some such result. But, even to secure Muslim unity, no
section of the Shar'i 'ulama.' — not even the Mu'tazilis, whom he favoured —
was prepared to abandon, in favour of caliphal power, the ideals of the old
Piety-minded opposition, even had this been feasible.

1 On the Arab and Khrasani troops under the early 'Abbasids, see Claude Cahen,
'Djaysh', in 2nd ed. of Encyclopaedia of Islam. H. A. R. Gibb, in 'The Caliphate and the
Arab States', in Kenneth M. Setton, ed., A History of the Crusades (University of
Pennsylvania Press, 1955), vol. 1, pp. 81-98, points out the change in the troops' out-
look. But the point of departure for al-Ma'mun's reign remains the study by Francesco
Gabrieli, Al-Ma'mun e gli 'Alidi (Leipzig, 1929).

DISSIPATION OF THE ABSOLUTIST TRADITION 479

The Fourth Fitnah: al-Ma'mun's Wars

809 Hariln al-Rashid dies, al-Amin becomes caliph according to the
'Meccan documents', al-Ma'mun establishes himself in Khurasan

810 al-Amin names son Musa in Friday prayers as successor,
bringing up question of al-Ma'mun's place in the succession
order as established in the 'Meccan documents'

8ro al-Ma'mun refuses to yield, is declared a rebel when he refuses

to come to Baghdad

811 al-Amin sends army under 'Alt b. Tsa against al-Ma'mun's
forces commanded by Tahir b. Husayn; 'Alt killed; Tahir defeats
a second army sent by al-Amin; al-Jibal province in hands of
Tahir's forces; Syria in disorder

812 al-Husayn, son of 'Alt, briefly deposes al-Amin and declares for
al-Ma'mun; al-Amin restored but al-Ma'mun's troops occupy
Khuzistan; Baghdad invested by al-Ma'mun's generals Tahir
and Harthamah b. A'yan

813 Tahir's men kill caliph when Baghdad falls; Harthamah had
promised him safe conduct

814-815 Abu-Saraya and Muhammad b. Ibrahim b. Tabataba raise

Shi'i revolt in Kufah, put down by Harthamah

816 Harthamah killed by al-Ma'mun's vizier al-Fadl b. Sahl

817 Baghdadis try to persuade Mansur, son of al-Mahdi, to declare
himself a claimant; he refuses

817 al-Ma'mun makes the Shi'i 'Ali al-Rida, ('eighth' imam) his heir

on the advice of his vizier Fadl, but Baghdadis and others in
the Iraq revolt; so also one of al-Ma'mun's generals

816-817 Babak raises a religio-social revolt

817 Baghdadis recognize Ibrahim, another son of al-Mahdi; 'Ali
al-Rida warns al-Ma'mun of the nature and gravity of the
situation; al-Ma'mun may have felt his vizier Fadl had not
faithfully so apprised him

818 'Ali al-Rida. dies; Fadl killed, possibly by order of caliph;
al-Ma'mun moves westward toward Baghdad

819 al-Ma'mun enters Baghdad; rival caliph submits; has then to
face a Khariji revolt, unrest in various provinces, and Babak's
revolt; Tahir too powerful to remove, becomes almost
independent governor in Khurasan, principal commander of
al-Ma'mun's troops

Soon after the Shi'i rebellion in the Iraq, al-Ma'mun had tried to capture
the Shi'i movement itself, naming as his own heir the imam of one of its
most popular sections, 'Ali al-Rida, grandson of Ja'far al-Sadiq. In theory,
this might have helped reconcile all Muslims; for Jama'i-Sunnis, in principle,
should accept as their own any successor designated by the received caliph.
The Shi'is proved capable of offering little real strength, however, and his
Shi'i turn added to the resistance of Baghdad. 'Alt al-Rida conveniently

480 THE CIVILIZATION OF THE HIGH CALIPHATE

died while al-Ma'mun was on his way to pacify Baghdad by making it his
capital, and the caliph made no further attempt to integrate Shi'i sentiment
into the regime so directly, though he continued to insist on general respect
for 'Ali.

He did take further measures, nevertheless, aimed at suppressing religious
factionalism. Without sponsoring any one theological school as such, he
rejected the more extreme formulations of the Hadith folk, who, as their
popularity at Baghdad increased, were becoming militant in their opposition
to those who disagreed with them, including the Shi'is. A sensitive Jama'i-
Sunni religious teacher like al-Muhasibi, condemned by the Hadithi Ibn-
Hanbal, scarcely dared teach publicly at the capital. Al-Ma'mun required
that all officials, qadis, and (so far as possible) 'ulama' generally reject
certain Hadithi doctrines, condemned by the Mu'tazilis (and by all other
specialists in kalam disputation at the time) and now also by the state, as
un-Islamic. But this meant adopting official theological positions and
enforcing conformity to them. To the extent that he succeeded, al-Ma'mun
not only countered religious factionalism and reduced the public power of
the Hadith folk ; more generally, he prepared the way for an Islamic institu-
tion more amenable than had been most of the earlier 'ulama' to the religious
demands of an absolute monarchy.

But this intervention was itself factional in effect. Of all the former
sections of the Piety-minded opposition, the Mu'tazilis had been the most
zealous in defence of Islam against its various non-Muslim opponents and
(perhaps partly for that reason) the readiest to rely on deductive reasoning
for their doctrine rather than on the letter of hadith reports. It was the
Mu'tazilis who had taken the lead in the campaign against the Manicheans.
They continued their severity against any un-Islamic infiltrations in official-
dom, but now turned their attention, with al-Ma'mun's aid and led by the
great qadi Ahmad b. Abi-Du'ad, to what seemed to be un-Islamic tendencies
within Shar'i Islam itself.

That section of the 'ulama' which specialized in gathering hadith reports —
whom we have called collectively the Hadith folk — had never been comfort-
able in too close contact with the worldly 'Abbasis. In addition, their own
special religious enthusiasm, one well adapted to private lives, went to
confirm their emotional independence of the existing Muslim state. It
simultaneously fired their zeal to impose their private factional norms
wherever they could, without regard to public policy.

As we have seen, this enthusiasm expressed itself in a doctrine that
identified the Qur'an very closely with God Himself: the Qur'anic words
which were so often piously upon their tongues were 'uncreated', unlike
everything else in the world. This conception, which stressed the immediate
accessibility of the divine law to every believer, was very appropriate to the
Shari'ah-minded conceptions of social equality and individual dignity; but
by the same token it was ill suited to the ideal of absolute monarchy. What

DISSIPATION OF THE ABSOLUTIST TRADITION 481

the Mu'tazili 'ulama' (and other men of kalam) disliked about it was perhaps
not so much the independent and factional spirit it fostered as the un-Qur'anic
illogic of a doctrine which seemed to idolize the Qur'an as almost a second
god, introducing a mystery into Islam contrary to the straightforward
reasonableness of the faith. But for both intellectual and political reasons,
al-Ma'mun was readily persuaded to decree that any who adopted the
doctrine that the Qur'an was uncreated were not true Muslims.

Those 'ulama' who dissented were, at best, excluded from public positions,
which tended to be filled with Mu'tazilis (many of whom, however, declined
to identify themselves with the regime even so). Thus the religious authority
of the state was staked upon the outcome of the Mu'tazilis' controversy
with the Hadith folk. But since the populace of the cities, especially Baghdad,
was turning against the Mu'tazilis, this meant basing the state religiously
rather on some of the intellectual elite than on the masses. It made for
disunity rather than unity. 2

Al-Mu'tasim: the dilemma of the personal army

Al-Ma'mun left to his brother al-Mu'tasim (833-842) the caliphate and his
controversy with the Hadith folk. Al-Mu'tasim continued his policies by
and large, though with less initiative and vigour. He made a notable departure
however, in military policy. If it was important to unite the empire on a
common basis for authority, one means to this seemed to be to give the
central power, the caliph, his own immediate base of military strength.
Unfortunately, this was achieved at the expense of weakening further such

2 Dominique Sourdel, 'La politique religieuse du caliphe 'abbaside al-Ma'mum',
Revue des Etudes Islamiques, 30 {1962), 27-48, is a very good study of al-Ma'mun's
policies. Sourdel stresses the Shi'J tendencies of some of the Mu'tazilis, and inclines
to see in the early Shi'i policy and the later Mu'tazili policy two forms of the same
impulse, which he associates with a Faylasuf-Mu'tazili-Shi'i party at court (to which
he prudently gives no name) set off against a Hadithi party; he suggests that under
al-Rashtd, the Faylasuf-Mu'tazili-Shi'i party had been favoured by the Barmakids,
while the Hadithis had been favoured personally by al-Rashid, and then by al-Amin
(and again by Ibrahim). I think that in his analysis of this supposed Faylasuf-Mu'tazili-
Shi'i party, however, he is misled by a premature retrojection of later sect lines back
into these formative times. He calls the Hadithi position then popular in Baghdad
'orthodoxy', and refers to its 'traditional' doctrines — confusing the two senses of the
Word — and hence misses its factionalist and innovative role; while he correspondingly
assumes that those who were then called 'Zaydis' held the full doctrine later elaborated
by al-Rassi ; and hence fails to bring out the ways in which al-Rida, against whom they
would have no immediate theoretical objection, might still be unacceptable to them.
Hence the point he does make, that al-Ma'mun was hoping to reconcile the more moderate
Jama'Ss and the more moderate Shl'is, is not sufficiently articulated. But it is far more
important than the hint of a Faylasuf-Mu'tazili-Shi'i coalition party — which may never
really have existed: without any such party, the intellectualist interests of the Barmakids
and of al-Ma'mun would have had like consequences independently in the fostering of
Falsafah and kalam discussion, in the conciliation of Shi'is, in the tolerance of dhimmi
non-Muslims — and in checking the Hadithi factionalism and communalism which
threatened all these policies.

482 THE CIVILIZATION OF THE HIGH CALIPHATE

links as existed between the caliphal power and the general Muslim populace.
The caliph's personal military guard, which had already included purchased
slaves in its ranks, was now put under the command of the caliph's personal
slaves rather than of free men of standing in the community. This move
might give the caliph independence from factional quarrels among the
Muslims and in particular from the dominance of the Khurasani troops (who
might be more loyal to their semi-independent Tahirid commanders than
to the caliph). It also gave the military guard, now a wholly slave corps
(and thousands strong), a high internal cohesion and a dangerous irrespon-
sibility vis-a-vis the wider community. The slaves were neither Arabic nor
Persian, but mostly Turks from the central Eurasian steppe (a fertile
ground for slave-raiding), who felt no tie with the local population in the
Iraq or any other province.

Their irresponsibly violent behaviour among the people of Baghdad
seems to have contributed to impelling al-Mu'tasim to a second move which
reinforced his alienation from the body of Muslims. He caused a new city,
Samarra, to be built (836) for governmental purposes along the Tigris a
good distance north of Baghdad, to which latter city he left the control of
commerce and scholarship. In Samarra, surrounded by his Turkic slave
guard, the caliph was indeed an absolute monarch, but one increasingly
out of touch with the socially dominant elements in the empire.

Al-Ma'mun had died during a defensive campaign against the Byzantines.
Al-Mu'tasim was able to renew the offensives of al-Rashid's time and carry
out the most ambitious raids since Marwani times. In Anatolia, Ankara
was wrecked and Amorium captured; preparations were made for another
siege of Constantinople, but when the Muslim fleet was destroyed in a
tempest, the project could not be carried through. Anatolia was again
abandoned. It was the last great foreign exploit of the caliphal state.

The project of creating a religious establishment tied to the government
meanwhile made little further progress. Religious communal sentiment,
indeed, had its way, and it even served to enforce cultural conformity.
Babak's Iranianizing rebellion in Azerbaijan gave occasion for sentiment at
the capital to harden against men who were sympathetic to the more
explicitly Iranian traditions. Victor (837) over Babak was al-Afshtn, who
was hereditary Persian ruler of a district beyond the Oxus, but also a
masterful general for the caliph. He had converted to Islam, but made
little secret of his sympathy for the old Iranian culture; he refused to be
circumcised, for instance, alleging fear for his health, and he loved the old
Pahlavi books with their richly colourful illustrations. On this basis, his
enemies at the capital were able to have him arraigned for treason, suggesting
that he had been secretly in league with Babak against the Muslims and the
Arab spirit Islam carried with it. In particular, they took note that in his
hereditary district he allowed himself to be addressed by titles that seemed
to imply divinity — a posture inconsistent with true Islam, He defended

DISSIPATION OF THE ABSOLUTIST TRADITION 483

himself with dignity; as to his title, in particular, he pointed out that the
continuity of his dynasty presupposed certain popular notions, which he
could not set aside without undermining the basis of his own rule (which
he had used, indeed, in loyal service to Islam). He himself laid no stock in
any such notions. But such politic arguments could not move an aroused
Muslim audience. His enemies had their way, and he was condemned to
death (840) ; his body, in the custom of the time, was shamefully desecrated.
The rebellions persisting from al-Ma'mun's time were finally crushed, and
under al-Mu'tasim's son and successor, al-Wathiq (842-847), the authority
of the caliphal government was not (in principle) seriously questioned in
the provinces (save always at the western end of the Mediterranean). But
the position of the Turkic slaves at the empire's heart became constantly
stronger.

The deterioration of the Sawdd irrigation

Apart from any political decline in the resources of the caliphal state in
human loyalties, after the reign of al-Ma'mun Baghdad was losing an
important economic basis on which its central political and cultural position
had been built. In the long run, economic failure may have been at least
as fateful for the absolutism, as the political failure with which it necessarily
interacted.

The rich Mesopotamian agricultural region, the Sawad, was returning
less revenue, not only to the government but in absolute yield. At least
in the Diyalah basin (the area nearest Baghdad), this resulted in part from
cumulative causes beyond governmental control. The land seems to have
been rising in such a way that the Diyalah river came to flow faster — and
sometime or other this increased rate passed a critical point: instead of
flooding periodically and depositing silt round about it, the river began to
scour out its own bed and lower its level as compared with the land around.
The high banks formed by the silt dropped nearest the river had presented
a relatively slight engineering problem for those who wanted to divert its
waters bit by bit for irrigation. But when the river bottom dug substantially
below the level of the plain, more had to be done than keep a way clear
through the high banks. For one thing, the engineers had to shift the canal
courses into more expensive patterns (eventually they even tried to pave
the river's bed to prevent further scouring). But for all the increased invest-
ment, less land was under cultivation.

But these geological changes (which were probably not limited to the
Diyalah) were not all. With the intensive irrigation that had been pushed
at least since Nushirvan's time, almost certainly large areas in the Sawad
must have succumbed to salinization — poor drainage producing a chronic
rising of the ground-water level to the point where its salty waters would
kill plant roots. Only a long process of restoration could normally drain out

t^

00

1

■<*■

p "

00_

< R

0i

W

P &

H

S i£

<<

►j

<

T3

S5

R

H oo

J? 1

nj

S> M

.«•

S o

3

|, °°

3

►J —

P
<

CO

<y oo
P o

<

*"> i

-O! M
3 3

S I

•< On

01 '•*

I
O

<M

O

0-

rt-

<•<!

Ch

»

1

J

ro

■<

O

S-t o>
■S5 I

2

&

DISSIPATION OF THE ABSOLUTIST TRADITION 485

such lands again. In any case, by the early tenth century the Sawad was
yielding a much smaller revenue than even a generation earlier. A point had
been reached at which the process of decline was self -accelerating. Military
turbulence, itself resulting largely from the reduced resources of a government
committed to high expenditures at court, contributed to hasten its own
causes: when, in 937, the great Nahrawan canal was breached by soldiers in
factional fighting between rival forces, it was not soon restored. Engineers
were ceasing, in any case, to attempt to maintain the old agricultural level.
Maintaining the Sawad had reached the point of diminishing returns and
no longer repaid the necessary investment.

In such circumstances, even civilian officials might give up the attempt
to maintain administrative standards; and once demoralization set in,
natural difficulties were accentuated by a tendency of officials to lower their
levels of expectation as to what constituted acceptable bureaucratic behaviour.
Corruption, always a danger, became worse.

But once the political and social structure of the Sawad had become
dependent on Baghdad and the other great cities, central intervention was
necessary if the area was not to decline agriculturally even below the level
which a more decentralized irrigation had maintained in pre-Sasanian
times. Local initiative was no longer in a position to restore a more moderate
prosperity, for the land was controlled from the cities; or else it readily
fell into the hands of pastoralists, who alone had the tribal social structure
and the nomadic alternative resources which would enable them to hold out
against pressure from the cities. In the following centuries, the Sawad was
eliminated as a primary source of centralized revenues in the region, giving
the government that controlled it resources to outlast rebellions and other
disruptions elsewhere in the empire. But there was no visible alternative
financial basis of the same order for a centralized bureaucratic empire. To
survive, the empire would have to find a more general economic basis for
continuity in its authority. The problem of religious legitimacy was in-
creasingly compounded by that of financial viability; and the political crisis
of confidence, by interrupting and demoralizing central control, in turn
hastened the economic troubles. 3

The crisis of caliphal absolutism

The next caliph after al-Wathiq, al-Mutawakkil (847-861), reversed the
religious policy of al-Ma'mun and al-Mu'tasim, but was incapable of bringing
the empire back to the non-committal but splendid strength of al-Rashid's
day. By his time, the Turkic troops were discovering that if they depended
on the caliph alone, so did he on them. The caliphs tried to offset such a

3 'Abd-al-'Aziz al-Durl (all of whose work is important) has studied economic history
at the end of the High Caliphal Period in Ta'rikh al-'Irdq al-iqlisddi fi'l-qarn al-rdbi'
(Baghdad, 1945)-

486 THE CIVILIZATION OF THE HIGH CALIPHATE

consequence by introducing several different ethnic elements in the troops,
whose rivalries would keep them all under control. Turks were set against
Negro slaves and against troops derived from the more westerly provinces.
But if, in fact, the soldierly forces were socially autonomous, then their
collective power rested, in effect, with whichever group among them could
gain the pre-eminence. Such a group could coerce the caliph without fear
of outside interference, once it had cowed its military rivals. The caliph's
efforts to divide and rule served his ends in the short run, but at any serious
crisis they merely resulted in more uncontrollable and destructive factional
fighting among the troops, making the commanders still less inclined to
behave responsibly.

Al-Mutawakkil was set on the throne in the first place at the will of the
Turkic guards. His personal life was taken up with trivial or vile extra-
vagances, some of which, no doubt, gave encouragement to the arts — this
was ideally one of the monarch's functions — but which in this case did so
at the expense of kingly responsibility. (For one party he expressed the
wish that all things should be yellow — guests must wear yellow clothes,
girls must be blondes, the food must be of yellow colour on golden dishes,
even the water flowing in a stream through the garden must be coloured
yellow with saffron. Someone underestimated the amount of saffron that
would be required before the caliph became too drunk to notice, so when
the saffron ran out, precious dyed stuffs had to be soaked out in the water
to keep its colour yellow.) His public policy was left to the most bigoted of
the Hadith folk, who celebrated, on the one hand, the end of the caliphal
attempt to control the Jama'i-Sunni 'ulama', while instituting, on the other
hand, active persecution of their opponents. The Mu'tazilis suffered disgrace,
but little more. But Shi'i shrines were obliterated. In accordance with the
doctrine that dhimmis might not build new places of worship after the
conquest, many such newly built churches and synagogues were torn down.

For fourteen years the government pursued no consistent imperial policy.
Then the one power close to the throne that possessed sufficient solidarity
to act effectively showed its hand: the Turkic slave soldiers murdered
al-Mutawakkil and freely installed his son in his place, unhindered by any
section of the public.

There followed a crisis in the state far more serious than that on al-
Rashid's death. At last the demand for Muslim unity, which had triumphed
readily in the time of Mu'awiyah and had persisted through every subsequent
crisis, proved relatively impotent. For ten years (861-870), a series of four
short-lived caliphs at Samarra tried vainly to evade the power of the Turkic
soldiers who made and unmade them. The first (al-Mutawakkil's son, al-
Muntasir) was branded as a parricide and was soon undone. One (al-Muhtadi)
stood out for his exemplary frugality and honesty, by which he hoped to
win religious awe and the support of the soberer of his subjects; he succeeded
in staving off crises by sheer bravado for several months before he, too,

DISSIPATION OF THE ABSOLUTIST TRADITION 487

succumbed. All were more or less creatures of one or another military
faction and necessarily bowed to the general in power, who had to loot the
treasury for his own troops and so prepare financially for his own fall. The
provinces meanwhile were largely left to their own devices.

The next serious caliphal ruler, al-Muwaffaq (870-891), brother and
lieutenant of the titular caliph al-Mu'tamid, succeeded in imposing his
authority on the troops by political address and by good luck. Long favoured
over other 'Abbasids by Baghdad opinion, he had to win his way by patiently
organizing military victories. He confronted rebellion and civil war in all
directions. In the west al-Muwaffaq permitted the Turkic governor of Egypt,
Ahmad b. Tulun, to be practically independent in Egypt and even to annex
Syria. (Ibn-Tulun then added to al-Muwaffaq's troubles by encouraging his
brother, the titular caliph, to flee al-Muwaffaq's frugal tutelage to Ibn-
Tuhin's protection, and the de facto ruler had to take out time to suppress
these intrigues. Still he preferred not to depose the disloyal brother but to
maintain a show of caliphal legitimacy.) In the north he could not prevent
the establishment of a Zaydi Shi'i state in Tabaristan, south of the Caspian.

In the east the governor of Sistan, Ya'qub b. al-Saffar, had risen from
frontier fighting as a local popular hero. Ya'qub had led local militia forces
against the last of the Khariji bands in eastern Iran, who had apparently
degenerated into robber bands; they could no longer win the passive support
of local populations, who were now turning Muslim themselves, in any case,
and felt a stake in the Muslim order. Ya'qub had risen rapidly to the governor-
ship of his province during the disturbances following al-Mutawakkil's death,
and gained the enthusiastic confidence of at least the city population. This
was partly due to his personal gifts as a captain, no doubt (it is said that he
first attracted public attention for his masterly management of a robber
band himself), but also to his social policy. An artisan's son himself, he led
forces the core of which seems to have been town-based and not at all
aristocratic in their loyalties. He and his successors of his family were
identified with the interests of the lesser townsmen as against the landed
gentry. In the general uncertainty, Ya'qub was persuaded to -widen the
sphere of his power; he led his men against the non-Muslim mountaineers
eastwards, and (867) set about a conquest of all Iran, where the local
populations proved not averse to his coming, especially as he seems to have
kept his troops under strict discipline. The aristocratic Tahirid regime
could not withstand him. Al-Muwaffaq defeated Ya'qub when he menaced
the Iraq, but had to allow Ya'qub and his brother, 'Amr, to retain an
independent governorship in most of Iran.

Al-Muwaffaq's main efforts were concentrated on the financially crucial
Iraq itself, which yielded less revenue than earlier but had not yet entered
on its last, most ruinous, phase. In the south, the Negro slaves, called
'Zanj', many of whom were used for labour in the marshy areas at the mouth
of the Tigris, had risen in 869 under a Khariji leader and set up their own

488 THE CIVILIZATION OF THE HIGH CALIPHATE

state, which tried to turn the tables on their former masters, enslaving the
former slave-owners. Basrah itself was sacked, and for a time much of
southern Iraq and Khuzistan was in their hands. They built themselves an
almost impregnable marsh city, which required special techniques for
al-Muwaffaq's soldiers to reach. It was only in 883 that the Zanj were
finally reduced.

Despite these interruptions, al-Muwaffaq was able to restore a degree of
strength and purpose to the caliphal state, though he did not overcome the
basic weaknesses that had characterized the reign of al-Mutawakkil. In his
hands and in the hands of his son al-Mu'tadid (891-902) and of his grandson
al-Muktafi (902-908), his triumphs bore fruit. Egypt was reoccupied when
Ibn-Tulun's sons proved incompetent. The resources that remained to the
central power proved sufficient for it to control the key block of provinces
extending from Egypt in the west to Fars in the east, from 'Uman in the
south to Armenia in the north. Revolts were endemic but were somehow
contained. More distant provinces in the west and south and east which
had been included in al-Rashid's empire retained full autonomy and con-
tributed little to the caliph's finances, but mostly acknowledged the caliphal
supremacy. A financial surplus was built up at Baghdad (which again became
the capital in 892). Yet the soldiery continued to form a potentially in-
dependent element of power even at the centre, though for the moment
subordinate to the civil administration.

The failure of the Qarmatian revolution

But the political initiative in Islamdom was shifting away from Baghdad —
whether Baghdad was in the hands of the caliph and his civilian bureaucrats
or of their military servants and occasional tyrants. Initiative lay increasingly
with those who could command the allegiance of the dominant elements in
the several provinces.

By the end of the ninth century, rebellions such as that of Babak, based
on an older national tradition, lost importance; so likewise did the earlier
type of general Khariji or Shi'i rebellion aimed at reforming the caliphal
state from within the established Muslim ruling classes. Neither a common
Muslim ruling class which felt itself set off as Muslim even if it included
many social levels, nor an identifiable and potent stratum of pre-Islamic
national leadership had retained political actuality: all major political
elements, at all the more active levels of society, were Muslims in common.
Instead of the earlier sorts of rebellion, the new movements which attacked
or at least detracted from the caliphal power tended to be based on blocs
of the population which could be assumed to be Muslim but whose political
ideas were limited to less comprehensive and amorphous interests than those
which now could concern the whole Muslim community.

The focus of attention came to be the province, such as Egypt or Syria or

X)
U

Hi

d
d
td

a

CI
h .d

d
a)
o

CD

<3

fa

Xf

§

+»

Oh

en

CD

bo
42

>

+3

'R

XI

cd

<nJ

d

45

u cd

&

3

CO MH

HH

t-i

•a °

cd

O

CD

1 e
"d 43
d d
d o

j_j CD

5 -a

XI

CO

o

6

o
cd

l

U
O

4H

N
'3

o

u
cd
P.

to

00

+> 0)

Oi

o

o ,
Oh iri

— d
0< 3

<B >cd

2 to

o

o 'u

+>.

o o

•2 |—j

4^

^a

co

CO

XI

!2 f -1 '

."S

> en
cd cd

'3

X) _u

W O" 1

S
<cd

<Cd CD

tS -d
cd d

<3

a

<cd

CD (0
fcH l-H

§3

H-

to- 3

W

cd OhN K

+>"

A

>,

so

W

£

o

p<

a

M

cd

+->

cd

o

u

o>

CD

o

O
-f-»

,g

CO

"3

•a

d

d

<3

oi

3

Oh

H

<u

.d

CO

x)
"P.

m

j3

to

d

<cd

■d

a

cd

J3

to

4)

6

Oh cd
cd 'm
Ai >,
cd to

o d
ii.5

a.

■d .-

rd CL

a w

o <3

M Oi

O ni

Oh

Oh<

ni

<oi
u
<oi

w

o

>

o

a

00

3

a

r3

H

en

^>!-

+J

3

O,

•'-p

O

a)

n

CO

OJ

•9

— H HOi

< a

d fe r rt
■3 g 3

3 2^

d

s

oi

PQ

4)
Oi

<3

PQ

be

ni

+j

x)

d

oi

oi
Oh

to

d

3

53 o

l.s

a> in

Oh Oh

"d ^

.a &
.a s?

oi g
Oh^

to P

d

CJ

13

d

u

M

Oh

CD

■d

fa

d

-H

oi

en

,_<

Xj

Oi

-H

cfi

'o

■d

t-H

ni
o

*— <

T)

H-H

^3

3

tO ^3

c?S

o> ^d '

eg

-p 43

d .

Oh <*
d ^

.a cs

43 i3

< s

_>,

"3

+-»

FH

d

(H

oi

3

>.

O

y

to

'C

*H-H

^

t—i

M-H

y

o

to

d

d

o

o

+3

3

X)

43

CD

a

cd

bo

"o

<J

CD

d

X)

OJ

ni

u

>

o

o

o

X)

3

t-

tn

S3

d

a

o

V.j*

0)
H- 1

>oS oi

cd

fa .£>

*

♦H CT"

.Oh

-d i

cd

tO >H

CJ

00
00

CN O
00 00

00

o

O^

■*

■o

Th

in

Oi

o

N

to

en

^*-

Tt-

CTv

Ov

Oi

Oi

Oi

Oi

ti

490 THE CIVILIZATION OF THE HIGH CALIPHATE

Fars, rather than the empire as a whole. Whatever esprit de corps there
was among the katib administrators of the caliphal state as a whole, it was
not strong enough to keep the local katibs from joining with the bourgeois
of their cities in their concern for the quality of the local governor and the
effects of his administration on the cities of the immediate region. The
sentiments of an older landed gentry — no longer the only military force of
the empire, and increasingly disrupted anyway by a tendency for the
administration to grant lands freely to new men — played at most a limited
role. When Ibn-Tulun had been able to withhold taxes from Baghdad he
kept them at home and spent them on public works in the cities of Egypt ;
this sort of change pleased the population and made them happy to see their
governor independent. Under these circumstances, little resistance would be
put up against either an appointed governor who threw off the caliph's yoke
or even a military adventurer, leading local warlike elements (in the Fertile
Crescent, for instance, Bedouin), who might take the governor's place.

The old-line Khariji and Shi'i movements had by now been restricted to
localized districts. Even the Hijaz took on the air, politically, of an isolated
province. The efforts that persisted there to establish local 'Alid rule were
eventually successful to a degree; but this had little meaning for any wider
Shi'i movement; the Hijaz was no longer a focus of any general political
movement. Yet Shi'ism had many adherents (notably among the Bedouin)
in those more central areas where al-Muwaffaq's campaigns had maintained
caliphal authority; discontent with an expensive caliphal government
perhaps could express itself in rejection of the Jama'i-Sunni 'ulama' who
.had compromised with it. But usually this was to mean only that some
locally-based governors and their supporters would be Shi'i in religious
sentiment, without carrying any distinctive Shi'i political programme.

One movement formed an exception, that of the Isma'ili Shi'is. It did
presuppose an all-Islamic framework, but not on the basis of the generally
recognized leaders of Muslim piety. It proposed to substitute at once a new
political and a new religious elite, and based itself, to this end, on mal-
contents of any class. As the Isma'ili faith was being spread by its often
heroic da'i missionaries, the ground was being laid on which full-scale
rebellion could flourish. Under the nickname 'Qarmatians', at the end of
the ninth century, an Isma'ili rebellion took place in the deserts between
the Iraq and Syria.

The rebels had a certain amount of peasant support (we read of chiliastic
peasant sects that identified their cause with the Qarmatians) ; and it must
have depended heavily on Bedouin, whom the caliphal state had dis-
illusioned long since. The leadership seems to have been urban and probably
won some lower-class support from the towns. A large number of young
men joined the Isma'ili camps in the deserts; they felt that the old way of
life was to be swept away, the privileged classes overthrown, and pure
justice was to reign. Tremendous bitterness was generated on both sides,

DISSIPATION OF THE ABSOLUTIST TRADITION 491

and the usual atrocities of civil war took place. We read of a pious mother
who went out to the desert camps to reclaim her son, who had joined the
rebels. She was horrified at the defiant atmosphere of the camps, egalitarian,
consciously rejecting the proprieties of established society — and demanding
rigid conformity to its revolutionary norms. The emancipated son made a
point of his toughness and cruelty, showing no acknowledgement of a
mother's dues ; the mother disowned him and returned full of denunciations
of the Qarmatians.

The movement continued for years to threaten 'Abbasi rule in the Iraq
itself. Any Shi'i might be accused of secret complicity in it, and jailed or
killed. But it was eventually suppressed ; the caliphal state was still strong
enough to maintain control of the more central provinces. This was the
empire's last major triumph.

Slightly further afield, the same name, 'Qarmatians', was applied to a
group in mainland Bahrayn, which had originally formed around a different
Shi'i faction but was persuaded (at least for a time) to accept the Isma'ili
cause. After the failure of the movement in the Syrian desert, this group
established its own republic in the east Arabian oasis towns. The govern-
ment was oligarchic and the state interfered directly to maintain conditions
of prosperity and equal economic opportunity in the towns; the forms of
Shar'i Islam were neglected, though not suppressed. The peasantry, however,
were not integrated into the new state but served merely, it would seem,
as resources for the townsmen. For a time, these Qarmatians sent expeditions
to distant points in Arabia and the Iraq (the most notorious such expedition
took Mecca and carried off the sacred Black Stone from the Ka'bah — these
Isma'llis regarded the reverence paid it as idolatrous; later it was returned).
They renewed some of the ferment of the Qarmatian revolt. But after
decades of quiescence, in the eleventh century, the Shari'ah was restored
in Bahrayn and the republican forms were destroyed, probably in part with
the help of the peasantry.

Fdfimis and Samanis

While the 'Abbasi caliphate did maintain a precarious hold on the inner
provinces, the outer provinces were being organized as autonomous Muslim
societies. There where there had been, at least since the death of al-Muta-
wakkil, no question of central interference, states could be constituted on
bases less ambitious than the overall restoration of the caliphal state. Even
the Isma'ili movement served more provincial interests there.

The most effective Isma'ili rebellion took place far from the centre, where
the caliphal power had long ceased to be effective. The Qarmatian movement
in the desert seems to have been associated with that section of the Isma'ili
leadership that made its headquarters in Syria. This leadership seems to
have proclaimed its own candidate to the imamate, who — whether the

492 THE CIVILIZATION OF THE HIGH CALIPHATE

actual descendant of Isma'ili or not — did not receive the allegiance of all
Isma'ili groups. Even the Qarmatians of the desert doubted him, and he
had to look elsewhere. His da'is prepared bases for him in both the Yemen
and the Maghrib; when they had won the support of a major Berber tribal
bloc (Sanhajah) in the interior of what is now Algeria, he travelled
thither (not without narrow escapes) and proclaimed himself true caliph
under the eschatological name al-Mahdi in 909; so founding what was
proud to be called the Fatimid dynasty, as descended from Muhammad's
daughter. Before enthusiasm waned, al-Mahdi succeeded in overthrowing
both the Ibadi Kharijis of Tahart and the Aghlabid dynasty of governors
in the eastern Maghrib.

The new caliph had to disown some of his followers who had been too
sanguine in their eschatological expectations, and threw in his lot with the
coastal merchant cities by moving the east Maghrib capital from Qayrawan,
oriented to the land routes of the desert, to a new foundation, Mahdiyyah,
a port. The change of atmosphere was signalized by the work of Qadi
Nu'man, a former Maliki legist who opportunely converted to Isma'ilism
and worked out a monumental system of Shar'i law, based on the school
claiming the authority of Ja'far al-Sadiq, in time to provide the new dynasty
with a working legal corpus of its own. It was explained that the eschatological
work of the Mahdi in subduing the world would be shared among several
rulers of his dynasty; meanwhile the state must function, on the whole,
like any other Muslim state. Al-Mahdi and his son (also called by an eschato-
logical title, al-Qd'im, 934-945) had to resist a vigorous Khariji Berber
revolt, which cooped the Isma'ilis up for a time in their new naval capital ;
but they had been able to win sufficient support in the cities to maintain
their organization intact and break the Berber resistance. With loyal Berber
troops, they conquered the whole of the Maghribi coast, subduing the last
of the Idrisid rulers in the Moroccan cities ; but they did not cross over into
Spain, which remained independent under its Umayyad amirs. Inheriting the
Aghlabi naval power, the Fatimids inherited also the Muslim position in Sicily,
and Mahdiyyah became a major commercial capital of the Mediterranean.

At the same time that the former Aghlabi domains were being taken over
by the Fatimid dynasty, the former Tahiri domains were being reintegrated
by governors of the Samanid family, the most loyal or at least respectful
to the caliphate of all the new powers that were arising. Governors at Bukhara
and Samarqand in the Zarafshan valley (north of the Oxus) since 875, in
903 (under Isma'il b. Nasr) they had taken Khurasan away from the Saffarid
family. Ultimately, social power lay with those who could rally the agrarian
forces, which the Saffarids had alienated. The Saffarids were soon driven
back to their home base in Sistan — where, however, they were able to
maintain their authority locally for many generations subsequently. The
Samani state thus constituted in northeastern Iran and the Syr-Oxus basin
was vast, but at least had in common a predominantly Persian population

DISSIPATION OF THE ABSOLUTIST TRADITION 493

and common aristocratic and military traditions. The dihqdns, landed gentry,
felt the state to be their own and defended it as giving national leadership
against the nomadic Turkic tribes. The struggle with the nomads had
coloured the whole aristocracy of the northeastern Iranians, exposed to
nomadic attack not just at a single frontier line but spottily throughout a
broad arid zone. This struggle was now identified with the defence at once
of Iranianism and of Islam. Under Nasr II (913-942), Samani rule brought
unquestioned peace and prosperity not only to the Oxus basin and Khurasan
but often to much of western Iran as well.

Within their realms, the Sa.ma.nids maintained the structure of the caliphal
bureaucracy almost intact, though modifying it somewhat according to local
practices ; they were enlightened patrons of literary culture both in Persian
and in Arabic. Under their rule Arabic adab cultivation found a congenial
home, almost as good as Baghdad. It was under them, also, that the new
Muslim Persian, adopted by the petty dihqan gentry, became a regular
literary language. Unlike the Saffarids, who had been closely allied with
less than aristocratic elements in Sistan, the Sa.ma.nids upheld the dignity
and the privileges of the gentry; correspondingly, they received the praise
of the lettered classes.

In the territories still effectively subject to the Baghdad caliphal govern-
ment, on the contrary, sound government — so far as it existed — was still
dependent on caliphal strength, and this was waning. At the succession to
al-Muktafi (908), it is said, ambitious courtiers deliberately preferred to
raise to the throne a boy who promised to be weak and manageable rather
than a grown man with ideas of his own (the adult candidate was Ibn-al-
Mu'tazz, the poet and critic). Al-Muqtadir, the boy, satisfied their short-
sighted hopes, if this was the case, only too well. His caliphate (908-932)
repeated the aimlessness of that of al-Mutawakkil, though on a somewhat
reduced scale. But now the caliphal state could not afford a weak reign.

The breakup

This time there was no recovery for the state. Not only the army was socially
rootless. Similarly, the state finances were commonly managed by wealthy
entrepreneurs (often Shi'is) interested in making a private profit, rather
than by an independent aristocracy or even by an entrenched self -perpetuating
bureaucracy, with a vested interest in governmental stability. What was left
of the old Sasanian bureaucratic class was perhaps becoming more attached
to local governors. Al-Muqtadir's court was wastefully extravagant; under
these circumstances, the result was financial mismanagement, ending in
chronic problems of fiscal supply and — when the soldiers' pay was con-
sequently slow — of military indiscipline. By the end of al-Muqtadir's reign,
the state was bankrupt. Simultaneously, the authority of Baghdad even in
the nearer provinces was rapidly shrinking. On al-Muqtadir's death (at the

494 THE CIVILIZATION OF THE HIGH CALIPHATE

unwilling hands of his best general, Mu'nis, driven to revolt for the sake of
the caliphal state itself), what was left of the empire quickly fell apart.

Four caliphs ruled briefly between 932 and 945, each of them at the mercy
of the military factions that had raised them to the throne. The most powerful
general came to be given, the title amir al-umard', commander-in-chief,
and was granted an overruling power in the state; the caliph gave up any
direct administrative role, retaining only ceremonial functions. But in any
case the government at Baghdad, usurped as it was by the soldiery, soon
ruled little more than the Iraq itself. The considerable territory which al-
Muqtadir had inherited was being divided among military powers less
responsible than those which now held the remoter western and eastern
provinces. Here Shi'i forces were prevailing, less spectacular and also less
ambitious than were the Isma'ilis, but also, for a time, more prominent in
imperial affairs.

Already by 897, Zaydi Shi'is had established their rule in the Yemen.
But this was a case apart, like the establishment of the Isma'ilis in the
Maghrib. Closer home, in 905 the Hamdanids, leaders of a Bedouin tribal
bloc, had established themselves as governors of Mosul in the Jazirah itself;
before the end of al-Muqtadir's reign, they became effectively autonomous.
They were Shi'is, like many of the northern Bedouin at the time, evidently
with Nusayrl leanings. But their main strength was their respected position
as chieftains among pastoralists, whom they could not displease with
impunity. From 944, a branch of the family was established at Aleppo.
There they had to bear the main burden of the Muslim resistance to a
renewed Byzantine military advance — almost the only non-Muslim attempt
on the lands of the caliphate during this period of weakening in the central
power. The Byzantines recovered Cilicia and much other territory just east
of the Anatolian highlands, and were driving into Syria; ghdzts volunteering
for the sacred war came from afar to stave off disaster. Sayf-al-dawlah, the
most famous of the Hamdanids, attracted poets and scholars to Aleppo:
he patronized al-Mutanabbi' and al-Farabi. He was less successful in the
war against Byzantium, in which he received little formal help from else-
where, and could not save Antioch; but the poets praised his valour none-
theless, and for his place in their songs he received as much renown as if
he had actually turned back the enemy.

By 928, in the lands southeast of the Caspian, the Ziyarids, military
leaders, had become independent and were absorbing the governorships of
'Iraq 'Ajami as well. Among their soldiers from Day lam, a small but rugged
sub-Caspian highland territory, three brothers, of the Buyid family (in
Arabic, Buwayhid), rose to high captaincies and soon made themselves
independent, with the help of Daylami troops, at once of the Ziyarids and
of the caliphs. By 934, they were in control of the greater part of western
Iran though, like the Hamdanids, in ruling it they could not ignore the
good pleasure of their men — professional soldiers, Daylamis and latterly

DISSIPATION OF THE ABSOLUTIST TRADITION 495

Turks also. The Buyids too were Shi'is, as were most Daylamis (they had
been converted largely by Shi'i rebels using their mountains as base). The
Buyids were inclined to the Twelvers, though not exclusively; they gave
official status in their territories to Shi'i holy days.

Not having risen, like the Tahirids or the Tulunids for instance, as regularly
appointed governors but as condottieri who were filling a political vacuum,
they had no special ties to the caliphs' government. Yet, like the Hamdanids,
they made no attempt actually to replace the caliphs with other supreme
figures: they merely nullified effectively their power and authority in what-
ever lands they could seize. They tried to carry on the caliphal administration
however, though in a simplified form.

Egypt, retaken from the Tulunids in 906, was held for the central power
through al-Muqtadir's reign, but in 937 the caliphal governor- — a captain
who had inherited from a state beyond Oxus the title Tkhshid' — proved
strong enough so that he could not be replaced. He proceeded to rule in-
dependently, controlling also much of Syria (and at first staving off the
Byzantines with more success than later did Sayf-al-dawlah) . His scions
reigned Actively under the powerful protection of Kafur, originally a Negro
slave-eunuch; only after his death could the Fatimids from Mahdiyyah
realize their long-sought goal of occupying Egypt.

The partisans of the universal absolute monarchy had failed to build a
viable power structure, having allowed the immense prestige of the caliphate
to be frittered away generation by generation. Nor had the Shar'i' 'ulama'
been able to provide a stable political order, either in co-operation with the
state under al-Ma'mun and al-Mu'tasim or in the considerable independence
of it which they enjoyed subsequently. But it was not merely the prestige
of the caliphate that was dissipated, but that of a monarchical tradition
which had begun long before Islam. It was the whole heritage of Irano-
Semitic absolutism that was in question; it was neither effectively supported
nor replaced.'*

In 945, one of the Buyid captains, whose brothers already controlled the
western Iranian highlands above the Iraq, occupied Baghdad. Despite his
Shi'ism, the Buyid chief at Baghdad (Mu'izz-al-dawlah) took the title
amir al-umard' , commander-in-chief, and acknowledged the theoretical
position of the 'Abbasid caliphs, who maintained a local court with con-
siderable local authority. But in fact the Iraq was now merely a province
under the new Buyid power. The caliphal state had ceased to exist as an
actual independent empire.

1 H. A. R. Gibb, in 'Government and Islam under the Early 'Abbasids: The Political
Collapse of Islam', in L' Elaboration de Vlslam (Paris, 1961), pp. 115-27, suggests that the
failure of the 'Abbasi state to maintain itself so long as the Sasanian was due to the lesser
relative strength of the agrarian classes in the 'Abbasi state, where the bureaucracy,
army, and religious institution all developed in relative independence of the monarchy.
I have rather followed this lead here. (The article also points to the remarkable prevalence
of 'Alid-loyalist sentiment in this period.)

A Selective Bibliography for Further Reading

The following works are arranged by subject, corresponding in part, but not
entirely, to the chapters in this work. I make no attempt at completeness,
selecting largely according to where I have a comment to make. I have
attempted only to point to works that form an effective starting point for
those who would like to follow up a given field in more detail. For a number
of more specialized topics, commented bibliography will be found at the
appropriate place in the footnotes ; this bibliography does not include all the
works cited in the text. Readers should refer to the Sauvaget-Cahen biblio-
graphy (cited below) before beginning further reading.

Works of general reference

The Encyclopaedia of Islam, 2nd ed. (E. J. Brill, Leiden, 1954-66), of which
the first two volumes and some following fascicles have appeared. Com-
posed of articles contributed by leading Islamicists. Each article includes a
bibliography. This is the basic work of reference on any subject connected
with Islam or Islamicate civilization, especially in pre-Modern times. For
articles dealing with religion and law in subjects not yet covered by the
2nd ed. see The Shorter Encyclopaedia of Islam, ed. Hamilton A. R. Gibb
and J. H. Kramers (E. J. Brill, Leiden, 1953), which has brought up to
date, especially in point of bibliography, selected articles from the first
edition of the Encyclopaedia of Islam (Brill, 1913-38), in four volumes
and a supplement, now largely out-dated but still to be referred to for
articles not found in the second edition or in the Shorter Encyclopaedia.

Historical Atlas of the Muslim Peoples, ed. R. Roolvink (Harvard University
Press, 1957). The only general historical atlas in the field; it has no index
and practically no historical detail, and it wants revision and more maps
of developments in the later pre-Modern periods. Harry W. Hazard,
Atlas of Islamic History (Princeton University Press, 1951) is in fact
primarily a study of the shifting power balance between Muslim and
Christian powers in the general region of the Mediterranean, century by
century; it does include some general information on political conditions
in various areas in each century, but almost nothing east of Iran or in the
farther north or south. A better atlas is promised in connection with the
new edition of the Encyclopaedia of Islam. Useful maps may be found in
G. LeStrange, Lands of the Eastern Caliphate: Mesopotamia, Persia, and
Central Asia from the Moslem Conquest to the Time of Timur (Cambridge
University Press, 1930), which also covers Anatolia; in the relevant articles
in the Encyclopaedia of Islam; in E. de Zambaur, Manuel de Genealogie et
de Chronologie pour I'histoire de V Islam (H. Lafaire, Hanover, 1927, reprinted
1955), which contains exhaustive tables of rulers; and in some of the
general secondary works noted below.

497

498 A SELECTIVE BIBLIOGRAPHY

Clifford E. Bosworth, The Islamic Dynasties: A Chronological and Genea-
logical Handbook (Edinburgh University Press, 1967), 'Islamic Surveys,
No. 5'. Contains brief notices of political events affecting the important
Muslim dynasties from the western Mediterranean through the Indie
region, and lists their rulers. Generally supersedes Stanley Lane-Poole,
The Mohammadan Dynasties . . . (Constable, London, 1893; reprinted
Paris, 1925), though the latter contains genealogical charts. E. de Zambaur,
mentioned above, in spite of some needed corrections, remains primary.

Bibliographies

W. A. C. H. Dobson, ed., A Select List of Books on the Civilizations of the
Orient (Oxford University Press, 1955). Useful for the general reader. The
sections on Islamdom (and India) list the best books for English-speaking
readers on most aspects of Islamicate civilization, including several trans-
lations.

Charles J. Adams, ed., A Reader's Guide to the Great Religions (The Free
Press, New York, 1965). Very good chapter on Islam including recent
developments.

Jean Sauvaget, Introduction to the History of the Muslim East: A Biblio-
graphical Guide, revised by Claude Cahen (The University of California
Press, 1965). Referred to as 'Sauvaget-Cahen'. The primary point of
departure for the student of the subject, though it almost omits lands east
of Persia, and in most cases gives more emphasis to political and social
history. The French edition (Adrien-Maisonneuve, Paris, 1961), though
not so recent, is yet useful.

J. D. Pearson, Oriental and Asian Bibliography (Crosby Lockwood, London,
1966). Not a bibliography in the usual sense, but usefully expands on some
aspects of Sauvaget-Cahen.

Bernard Lewis, 'The Muslim World', in American Historical Association,
Guide to Historical Literature (Macmillan, New York, 1961). Good intro-
duction to pre-1450 period studies for the non-specialist historian. In the
same volume, Roderic Davison, 'The Middle East since 1450', is weighted
toward political interests. The pages on Islam under 'History of Religions'
in the same volume are of little use.

D. Gustav Pfannmuler, Handbuch der I slam-Liter atur (de Gruyter, Berlin,
1923). Though old, still useful on religious matters; gives contents and
conclusions of the writers listed. To be consulted for the history of nine-
teenth and early twentieth century Western scholarship on any given
point.

J. D. Pearson, Index Islamicus, 1906-1955 (Heffer, Cambridge, 1958). An
exhaustive, uncommentated listing of articles in European languages on
Islamic subjects in periodicals and other collective publications, by subject,

A SELECTIVE BIBLIOGRAPHY 499

with cross-references, for tracking down details. Supplementary volumes
appear regularly.

Denis Sinor, Introduction a I'etude de I'Eurasie centrale (Harrassowitz,
Wiesbaden, 1963). Detailed annotated bibliography, linguistic and
historical, on the Altaic and Finno-Ugrian peoples, especially Turkic (but
not on such major bodies, distant from the Steppe, as Ottomans and
Hungarians).

Journals

FOR THE GENERAL READER

The Middle East Journal, published by the Middle East Institute,
Washington, D.C. Concentrates on modern matters, including a review of
current events from Morocco to Bengal; also has reviews and notices of
books and periodical articles, and a list of scholarly journals.

The Muslim World, published by the Hartford Seminary Foundation,
Hartford, Conn. Primarily designed to give Christian missionaries in-
formed insight into Islam, but its range of interest is very wide ; also has
reviews and notices.

The Islamic Quarterly, published by the Islamic Cultural Centre, London.
Edited by Muslim missionaries for English-speakers, it carries articles by
Muslims and by non-Muslim Islamists, sometimes of high quality.

FOR CURRENT SCHOLARSHIP

Revue des etudes islamiques, including its Abstracta Islamica which reviews
both books and articles, including those in non- Western languages.

Bulletin of the School of Oriental and African Studies (University of London),
which has a very high proportion of important Islamicate historical
material.

Studia Islamica, appearing irregularly with high-quality studies, in French

and English.
Arabica, very good material in the field of Arabic studies.

For other excellent journals, see Sauvaget-Cahen, chapter 10.

On the historical development of the Islamicate lands generally :

Felix M. Pareja-Casanas, Islamologia (Orbis Catholicus, Rome, 1951 ; editions
in Spanish and in Italian). A collection of information on most aspects of
Islamicate history and culture, with extensive but unannotated biblio-
graphies.

Bertold Spuler, Geschichte der Islamischen Lander, 3 vols. (Brill, Leiden,
1952-59; in the series Handbuch der Orientalistik, ed. Bertold Spuler).
The first two volumes translated into English by Frank R. C. Bagley (and

500 A SELECTIVE BIBLIOGRAPHY

corrected) as The Muslim World: A Historical Survey, Parts I and II (E. J.
Brill, Leiden, i960). A summary of some political events with biblio-
graphy.

Carl Brockelmann, Geschichle der Islamischen V biker und Staaten (Munich,
1939). Translated into English by Joel Carmichael and Moshe Perlmann,
History of the Islamic Peoples (Routledge and Kegan Paul, London, 1949) .
A densely written political narrative including some cultural history;
largely on the Ottoman Empire; ignores the Islamicate peoples east of
Persia.

Sydney N. Fisher, The Middle East, a History (Knopf, New York, 1959).
Chiefly on the Ottoman Empire and its successor states ; readable but not
fully reliable.

V. V. Barthold, Mussulman Culture [1918], badly translated from Russian by
Shahid Suhrawardy (The University of Calcutta, 1934). A brief and
brilliant historical review from a Central Eurasian standpoint.

F. August Miiller, Der Islam im Morgen- und Abendland, 2 vols. (Berlin,
1885-87). A narrative, carrying the story through the fifteenth century;
it is now superseded by more recent studies, except that its description of
'Abbasid period developments is still useful.

On the Islamic religion generally:

Hamilton A. R. Gibb, Mohammedanism, an Historical Survey [1949], 2nd ed.
(Oxford University Press, 1953). Outstanding brief study of the religion as
it was formed in the earlier generations and evolved later.

Ignaz Goldziher, Volesungen iiber den Islam (Carl Winter, Heidelberg, 1910) ;
French translation by Felix Arin as Le Dogme et la loi de V Islam (Paul
Geuthner, Paris, 1920). Equally brilliant with Gibb (and covering the
same material), and more detailed, with useful notes; but somewhat out
of date.

Kenneth Morgan, ed., Islam — the Straight Path : Islam Interpreted by Muslims
(Ronald Press, New York, 1958). Prominent Muslim theologians and
scholars acquainted with Western intellectual life interpret, for Westerners,
Islam as they see it in their several parts of the world. Repetitive and often
superficial.

Thomas Arnold, The Preaching of Islam, 2nd ed. (Constable, London, 1913) ;
the edition of 1896 is much less full. Primarily a study of the methods used
for spreading the Muslim faith, it is now rather out of date ; but at least it
chronicles the expansion of Islam from the beginning through the nine-
teenth century in all parts of the world.

Kenneth Cragg, The Call of the Minaret (Oxford University Press, 1956). The
first half is a sensitive interpretation of the Muslim faith by a Christian

A SELECTIVE BIBLIOGRAPHY 501

missionary with true respect for Islam; the latter half discusses the task
of the Christian mission among Muslims as he sees it.

General works on Islamicate culture under the High Caliphate:

Bernard Lewis, The Arabs in History [1950], 3rd ed. (Hutchinson's Univer-
sity Library, London, 1956). A brief and incisive historical survey of social
and political history, especiallv before 1000 ce; since in this period
Arabic and Islamic history largely coincided, in chapters 2-8 it covers
Islamic society as a whole.

Philip K. Hitti, History of the Arabs [1937], 8th ed. (Macmillan, London and
New York, 1964). Lengthier than Lewis, but undependable in its guiding
conceptions and in its emphases. Far more fair-minded than Muir, cited
below. The work may best be used for some detailed facts on pre-Islamic
Arabia and on political developments up to about the year 750 ce.

Maurice Gaudefroy-Demombynes (and Sergei F. Platonov), Le Monde
musultnan et byzantin jusqu'aux Croisades (Boccard, Paris, 1931); in
Histoire du Monde, ed. E. Cavaignac, vol. VIII. Depicts the changing
political institutions more than the course of events, and is mostly super-
seded now by

Edouard Perroy, ed., Le Moyen Age; V expansion de I'orient et la naissance
de la civilisation occidentale (Presses universitaires de France, Paris, 1955) ;
in Histoire generate des civilisations, ed. M. Crouzet, vol. III. Includes the
early Ottoman period, and remarks some important social and economic
changes.

Thomas Arnold and Alfred Guillaume, eds., The Legacy of Islam (Oxford
University Press, 1931; and reprs.). Surveys aspects of Islamic cultural
developments in the first four to five centuries, mainly from the point of
view of their legacy to later European developments.

Gustave E. von Grunebaum, Medieval Islam [1946], 2nd ed. (University of
Chicago Press, 1953). Traces the outcome, in the vicissitudes of history,
of the world view initiated with Islam. Primarily on the High Caliphal
period, but brings in some later material without special notice.

Gustave E. von Grunebaum, Islam: Essays in the Nature and Growth of a
Cultural Tradition [1955], 2nd ed. (Routledge and Kegan Paul, London,
1961). A major supplement to the above.

Alfred von Kremer, Culturgeschichte des Orients unter den Chalifen, 2 vols.
(W. Braunmuller, Vienna, 1875-77). Old but interesting. Partial translation
by S. Khuda-Bukhsh as The Orient under the Caliphs (University of
Calcutta, 1920).

Eric Schroeder, Muhammad's People: A Tale by Anthology (Bond Wheel-
wright Co., Portland, Maine, 1955). A beautiful sequence of translations
from Arabic, pieced together to tell the story of the Muslims to about

502 A SELECTIVE BIBLIOGRAPHY

iooo ce. A bit romanced and undifferentiated, it reflects traditional
judgments (e.g. on Umayyads, Mu'tazilis) subject to revision.

On some 'non-Arabic' pre-Islamic influences affecting Islam and Islamicate
civilization:

Jorg Kraemer, Das Problem der islamischen Kulturgeschichte (Niemeyer,
Tubingen, 1959)- The booklet itself is a general essay, but its notes provide
an introduction to the studies (mostly scattered in articles) on the transi-
tion from various pre-Islamic cultures into Islamicate culture, especially
from Hellenistic.

Tor Andrae, Der Ursprung des Islams und das Christentum (Kyrkohistorisk
arsskrift, Vols. 23-25, Uppsala and Stockholm, 1926). Later studies have
cast doubt on such points as the relation of Christian sermons to the
Qur'anic surahs.

Richard Bell, The Origin of Islam in Its Christian Environment (Macmillan,
London, 1926). On Christian influences on Muhammad, stressing Muham-
mad's independence.

Abraham Geiger, Was hat Mohammed aus dem Judenthume aufgenommen?
[ J 833]; translated into English by F. M. Young, Judaism and Islam
(U.D.C.S.P.C.K., Madras, 1898). Only partially superseded by

Abraham I. Katsh, Judaism in Islam (New York University Press, 1954).
Studies of the Jewish origin of Qur'anic material. Unfortunately, two
curious studies of Jewish influence on Muhammad are based on unwarranted
assumptions : Charles C. Torrey, The Jewish Foundation of Islam (Jewish
Institute of Religion, New York, 1933) ; and Hanna Zakarias, De Moise a
Mohammed (Minard, Paris, 1955).

On Muhammad's life and times :

Frants P. W. Buhl, Muhammeds Liv [1903] ; translated by Hans H. Schaeder,
Das Leben Muhammeds (Quelle und Meher, Leipzig, 1930). Still a sound,
full-scale narrative biography.

Tor Andrae, Mohammed, sein Leben und sein Glaube (Vandenhoeck und
Ruprecht, Gottingen, 1932). English translation by Theophil Menzel,
Mohammed, the Man and His Faith (George Allen and Unwin, London,
1936). Brief, masterly interpretation of the more inward side of Muham-
mad's life.

William Montgomery Watt, Muhammad at Mecca and Muhammad at
Medina (Oxford University Press, 1953 and 1956). A detailed study,
bringing out the social and political implications of Muhammad's work
and statesmanship. Modifies Buhl's conclusions and complements Andrae.

A SELECTIVE BIBLIOGRAPHY 503

Wm. M. Watt, Muhammad: Prophet and Statesman (Oxford University
Press, 1961). A brief digest of the above.

Regis Blachere, Le Probleme de Mahomet, Essai de biographie critique du
fondateur de I' Islam (Presses universitaires de France, Paris, 1952).
Stresses how much we still do not know.

ibn Ishaq, Life of Muhammad, translated by Alfred Guillaume (Oxford
University Press, 1955); ibn Hisham's additions have been placed awk-
wardly at the rear. The earliest extant biography, generally acknow-
ledged by Muslims as authoritative.

Leone Caetani, Annali dell' Islam, 10 vols, in 12 (U. Hoepli, Milan and Rome,
1905-1926). The first forty years after the Hijrah are examined in detail
but not always profoundly, with the chief sources translated. Caetani's
evaluations are not always sound.

Translations of the Qur'dn into English:

The best translation for overall effect in continuous, careful reading is
probably Arthur J. Arberry, The Koran Interpreted, 2 vols. (George Allen
and Unwin, London, 1955). It makes use of archaic language which some-
times miscarries. Richard Bell, The Qur'dn, 2 vols. (T. and T. Clark,
Edinburgh, 1937), is more exact but not designed for continuous reading
and is sometimes marred by tendentious interpretations; the traditional
order of verses is rearranged. N. J. Dawood, The Koran (Penguin Books,
1956) is perhaps the most readable translation but is very free and all too
often arbitrary in its renderings. Mohammed M. Pickthall, The Meaning
of the Glorious Koran [1930] (now in Mentor paperback) is best avoided
because of its inept archaizing and its dubious modernizing interpretations.

Probably the best alternative to Arberry remains Edward H. Palmer, The
Koran [1880] (now in World's Classics, Oxford University Press). It is
less readable than Dawood but much more accurate. Still less readable
than Dawood, but also more accurate, is John M. Rodwell, The Koran
[1861] (now in Everyman's edition); like Bell's translation, Rodwell's
has the verses rearranged. There are also several English translations by
modern Indian Muslims, in which the slant of a particular brand of Islam
is often visible. Abdullah Yusuf 'Ali, The Holy Qur'dn (Muhammad
Ashraf, Lahore, 1938) is not easily disengaged from its modernizing com-
mentary. Muhammad 'Ali, The Holy Qur'dn, 4th ed. (Ahmadiyyah Press,
Lahore, 195 1) is unusually precise, and his Ahmadi doctrine is chiefly, but
not entirely, confined to the footnotes. Finally, the translation of George
Sale (1734) (Wisdom of the East series) was for a century and a half the
standard English version; it is a sound rendering, from the point of view
of the Sunni tradition, but heavy, and interwoven with the traditional
Sunnl exegesis — which is an advantage if that is what one wants. Its
footnotes summarize the most famous Sunni commentaries.

504 A SELECTIVE BIBLIOGRAPHY

Regis Blachere, Le Cor an; traduction selon un essai de reclassement des
sourates, 2 vols. {G. P. Maisonneuve, Paris, 1947-50) is a more compre-
hensive and judicious presentation of the Qur'an as known to early Islam
than anything in English; Vol. I (1947) is a historical introduction, a far
sounder guide to the student than Richard Bell's Introduction to the
Qur'an (Edinburgh University Press, 1953), which is important but highly
conjectural. Blachere's Introduction au Coran has been reworked for in-
dependent printing (1959) and his translation, Le Coran, has been reduced
from two volumes to one, retaining the more essential notes (1957).

On the early Caliphate and through Marwani times:

There is still no serious modern account of the early Arab conquests, in spite
of one or two recently published works. For Syria, Egypt, and westward,
see The Cambridge Medieval History, Vol. II, Chapters XI and XII,
written by Carl H. Becker. For Transoxania, see Hamilton A. R. Gibb,
The Arab Conquests in Central Asia (Royal Asiatic Socety, London, 1923).

Julius Wellhausen, Prolegomena zur dltesten Geschichte des Islams, in Skizzen
und Vorarbeiten, VI (G. Reimer, Berlin, 1899), and Das arabische Reich
und sein Sturz {G. Reimer, Berlin, 1902), the latter translated by Margaret
Weir, The Arab Kingdom and Its Fall (University of Calcutta, 1927, and
repr. Khayat's, Beirut, 1963). Still the standard account.

Leone Caetani, Annali dell'Islam (see above) is useful for the period before
Mu'awiyah.

Julius Wellhausen, Die religios-politischen Oppositionsparteien im alien
Islam (Weidmann, Berlin, 1901). Still basic on the political history of the
early Shi'is and Kharijis.

Henri Lammens published a number of detailed studies of the Umayyad
period; they are noted by Sauvaget-Cahen. Lammens sometimes reached
too far, but his work is indispensable; that of Francesco Gabrieli (noted
also in Sauvaget-Cahen) is surer but more pedestrian.

Wm. Muir, The Caliphate, Its Rise, Decline, and Fall, revised by T. H. Weir
(J. Grant, Edinburgh, 1915). Follows the traditional Arab accounts, with
revisions after Wellhausen and others; has a nineteenth-century Christian
bias.

The studies of Hamilton A. R. Gibb on this as on other periods are important-
Some are collected in Studies on- the Civilization of Islam, Stanford J.
Shaw and William R. Polk, eds. (Beacon Press, Boston, 1962).

Ahmad al-Baladhurt, Kitdb futuh al-bulddn, translated as The Origins of the
Islamic State, Vol. I by Philip K. Hitti (Columbia University, New York,
1916), Vol. II by Francis C. Murgotten (Columbia University, New York,
1924). The work deals with the campaigns of conquest and with the

A SELECTIVE BIBLIOGRAPHY 505

political and financial dispositions of each territory which came under
Muslim rule. The translations are slightly abridged or expurgated.

On political and administrative developments under the High Caliphate:

William Muir, The Caliphate, revised by T. H. Weir (see above), is available
for the earlier part of this period also.

Dominique Sourdel, Le vizirat 'abbdside de 749 a 936, 2 vols. (Institut
francais de Damas, 1959-60). Sticks close to its subject, but illuminates
the development of the administration. Extensive bibliography on the
period and commentary on the sources.

Thomas Arnold, The Caliphate (Oxford University Press, 1924). Now no
longer very useful, but readably made the point that the Caliphate was
not a Muslim counterpart to the Papacy. On the Caliphate as an insti-
tution, see M. Khadduri and H. J. Liebesney, Law in the Middle East,
Vol. I (Middle East Institute, Washington, D.C., 1955), Chapter 1, written
by H. A. R. Gibb.

Emile Tyan, Institutions du droit public musulman, 2 vols. (Recueil Sirey,
Paris, 1954-57). Describes the governmental institutions mainly theore-
tically ; must be used with care.

Harold Bowen, The Life and Times of ' Ali b. 'Isa, the Good Vizier (Cambridge
University Press, 1928). A close study of the vizier's functions in the early
tenth century ce.

Miskawayh and others [Portion of the 'Experiences of the Nations'], trans-
lated by H. S. Amedroz and D. S. Margoliouth as The Eclipse of the Abbasid
Caliphate (B. Blackwell, Oxford, 1920-21). Volumes 4 to 7 give the trans-
lation of this and other Arabic historians of the tenth-eleventh centuries.

On general aspects of social and cultural life under the High Caliphate:

Adam Mez, Die Renaissance des Islams (C. Winter, Heidelberg, 1922);
translated by S. Khuda-Bukhsh and D. S. Margoliouth, The Renaissance of
Islam (Luzac, London, 1937). A bad translation of a work good for detail
on tenth century Islamdom and for colour, but weak in its generalizations
and in its proportions, and biased by nineteenth century prejudices.

Bertold Spuler, Iran in friih-islamischer Zeit: Politik, Kultur, Verwaltung,
und offentliches Leben . . . 633 bis 1055 (Franz Steiner, Wiesbaden, 1952).
A great deal of information without deep insight.

Nabia Abbott, Two Queens of Baghdad: Mother and Wife ofHdrun al-Rashid
(University of Chicago Press, 1946). A careful compendium.

H. G. Farmer, A History of Arabian Music to the Thirteenth Century (Luzac,

506 A SELECTIVE BIBLIOGRAPHY

London, 1929). By 'Arabian' he means 'Arabic' and even 'Islamicate'.
Largely on social and literary aspects.

E. Levi-Provencal, Histoire de I'Espagne Musulmane, 3 vols. (Maisonneuve,
Paris, 1950-53). The definitive study of political history and culture there
to the eleventh century ce.

Antoine Fattal, Le statut legal des non-musulmans en pays d'Islam (Im-
primerie catholique, Beirut, 1958). Comprehensive; based mainly on theor-
etical sources, though includes some historical fact, especially in the Arab
countries ; stops before the later Middle Ages.

Solomon D. Goitein, Studies in Islamic History and Institutions (E. J. Brill,
Leiden, 1966). An uneven collection of articles, of which several survey
some social and economic developments.

Solomon D. Goitein, Jews and Arabs, Their Contacts through the Ages
(Schocken, New York, 1955). A popular review of their contacts through-
out history.

Walter J. Fischel, Jews in the Economic and Political Life of Mediaeval Islam
(Royal Asiatic Society, London, 1937). Studies five instances where Jews
became prominent (three in the High Caliphal period), bringing out the
social implications of their careers.

On classical Arabic literature:

Hamilton A. R. Gibb, Arabic Literature: An Introduction (Oxford University

Press, 1926, 2nd ed., 1963). A masterly brief survey of the pre-Modern

literature, noting good translations.
Reynold A. Nicholson, A Literary History of the Arabs (T. Fisher Unwin,

London, 1907; Cambridge University Press, 1930 and later). An older but

lengthier survey than Gibb's with short translations; mainly useful for the

earlier centuries.
Gustav E. von Grunebaum, Die Wirklichkeitweite der friiarabischen Dichtung;

eine liter aturwissenschaftliche Untersuchung (Vienna, 1937). Suggests the

limits and consistency of the worldview of pre-Islamic Arabic poetry.
Charles J. Lyall, Translations of Ancient Arabian Poetry (Williams and

Norgate, London, 1885 ; repr. 1930). Chiefly pre-Islamic poetry, with an

excellent introduction and notes; his translations are at least as faithful

as any others.
Arthur J. Arberry, The Seven Odes: The First Chapter in Arabic Literature

(George Allen and Unwin, London, 1957). Modern, almost imagistic

translations of the seven celebrated pre-Islamic 'odes'.
'Amr al-Jahiz (Gahiz), Le livre des avares, translated by Charles Pellat ( G.P.

Maisonneuve, Paris, 1951). A sample of the best work of this entertaining

writer.

A SELECTIVE BIBLIOGRAPHY 507

Gustave von Grunebaum, A Tenth-Century Document of Arabic Literary
Theory and Criticism: The Sections on Poetry of al-B&qilldni's I'jdz al-
Qur'dn (University of Chicago Press, 1950). An influential theologian
compares the Qur'an and Imru'al-Qays in point of verbal beauty.

Ibn-Kutaibah, Introduction au livre de la poesie et des poetes, translated by
Maurice Gaudefroy-Demombynes (Societe . . . Les Belles Lettres, Paris,
1947). A ninth century ce Arab literary critic's remarks.

A number of anthologies exist. Valuable are Reynold A. Nicholson, Transla-
tions of Eastern Poetry and Prose (Cambridge University Press, 1922);
representative writings in Arabic and also Persian. Arthur J. Arberry,
Aspects of Islamic Civilization as Depicted in the Original Texts (George
Allen and Unwin, London, 1964) ; also Persian as well as Arabic examples.
Eric Schroeder, Muhammad's People; a tale by anthology (see above).

On the social outlook of Shari' ah-minded Islam:

Noel J. Coulson, A History of Islamic Law (Edinburgh University Press, 1964),

'Islamic Surveys No. 2'. An excellent brief introduction to the Shari'ah

and to modern developments.
Joseph Schacht, An Introduction to Islamic Law (Oxford University Press,

1964). An alternative to the above.
Maurice Gaudefroy-Demombynes, Muslim Institutions, translated by John

MacGregor (George Allen and Unwin, London, 1950). Suggests some social

implications of Shar'i principles, and describes in its chapter on The Cult,

the salat, the Hajj, and other Muslim rites.
Duncan B. Macdonald, Development of Muslim Theology, Jurisprudence, and

Constitutional Theory (Scribner's, New York, 1903; repr. Khayats', Beirut,

1965). Somewhat outdated but still illuminating.
Louis Gardet, La cite musulmane, vie sociale et politique (Vrin, Paris, 1954).

The pre-Modern Shari'ah-minded social ideal studied with insight;

written from a Catholic viewpoint, with an eye to implications for modern

Muslim peoples, especially the Arabs.
Alfred Guillaume, The Traditions of Islam (Oxford University Press, 1924).

Not on the traditions, in the normal English sense, but on the hadith

reports and the system of their criticism ; based on
Ignaz Goldziher, Muhammedanische Studien, Vol. II (Niemeyer, Halle, 1890);

edited and translated by C. R. Barber and S. M. Stern as Muslim Studies,

Vol. II (Aldine, Chicago, 1966). The classic study of the growth of the

hadith.
Joseph Schacht, The Origins of Muhammadan Jurisprudence (Oxford Univer-
sity Press, 1950) . On the development of fiqh and of such hadith as were

used therein, showing the occasion for and the effects of the disputes over

ra'y, qiyds, etc.

508 A SELECTIVE BIBLIOGRAPHY

Majid Khadduri and Herbert J. Liebesny, eds., Law in the Middle East,
Vol. I, 'Origin and Development of Islamic Law' (Middle East Institute,
Washington, 1955). The first four chapters are magnificent presentations
of the formation of the Shari'ah and of its basic theories; the remainder
discusses the provisions of most aspects of the law that would come be-
fore the courts, with special reference to the Arab lands and the Ottoman
empire.

Asaf A. A. Fyzee, Outlines of Muhammadan Law (Oxford University Press,
1949) gives more details of the law as applied in modern India.

'Ali al-Mawardi, Les statuts gouvernementaux, translated by E. Fagnan
(Jourdain, Algiers, 1915). Outlines the classical Shar'i tradition, modified
with regard to the problems posed by the downfall of the caliph's power
in the tenth century.

al-Bukhari, al-Sahih' translated by 0. Houdas and William Marcais as Les
Traditions islamiques, 4 vols. (Imprimerie nationale, Paris, 1903-14). The
prime canonical collection of hadith reports.

al-Muhaqqiq al-Hilli, Shard'i' al-Isldm, translated by A. Querry as Droit
musulman, 2 vols. (Imprimerie nationale, Paris, 1871-72). The favourite
book of fiqh among Twelver Shi'i Muslims.

On devotional life in High Caliphal times and after:

Duncan B. Macdonald, The Religious Attitude and Life in Islam (University
of Chicago Press, 1909; repr. Khayat's, Beirut, 1965). Still a good intro-
duction.

Reynold A. Nicholson, The Mystics of Islam (George Bell, London, 1914;
repr. by Routledge and Kegan Paul, London, 1963). The most elementary
of his several good studies. A good introduction, alternatively with

Arthur J. Arberry, Sufism, an Account of the Mystics of Islam (George Allen
and Unwin, London, 1950). Less perceptive than the above, but gives
many translated excerpts.

G. C. Anawati and Louis Gardet, Mystique musulmane: aspects et tendances,
experiences et techniques (Vrin, Paris, 1961). An insightful survey of earlier
Sufism, with an analysis of the classical practices from a Catholic view-
point willing to see in them not only 'natural' but even 'supernatural'
mysticism.

Louis Massignon, Essai sur les origines du lexique technique de la mystique
musulmane [1922], 2nd ed. (Vrin,- Paris, 1954). An intensive study of the
development of Sufi thought within the Muslim tradition.

Louis Massignon, La passion d'al-Hosayn ibn Mansour al-Hallaj, martyr
mystique de V Islam (P. Geuthner, Paris, 1922). Studies in detail the most
famous case where Sufism impinged on the public consciousness, and all
the ramifications of it; a magnificently subtle study.

A SELECTIVE BIBLIOGRAPHY 509

Margaret Smith, Readings from the Mystics of Islam (Luzac, London, 1950).
Very brief selections translated. Her other, more extensive studies are
also sound.

'Alt Tabari, 77^ Book of Religion and Empire, translated by A. Mingana
(Manchester University Press, 1922). By a Christian convert to Islam, this
work {patronized by a caliph) defends Islam against Christian attacks;
presents Muslim piety on a relatively high level.

Dwight M. Donaldson, The Shi'ite Religion, a History of Islam in Persia and
Irak (Luzac, London, 1933). A quite uncritical presentation of later
orthodox Imami views about the early Shi' ah, which does bring out much
of the feeling in its legends and doctrines. It can be corrected in part by
Marshall G. S. Hodgson, 'How Did the Early Shi'a Become Sectarian',
Journal of the American Oriental Society, Vol. 75 (1955), pp. 1-13.

On philosophy, science, and theology in High Caliphal times and after:

Henri Corbin, Histoire de la philosophic islamique, Vol. I, with others to follow
(Gallimard, Paris, 1964). The best brief introduction to the whole field;
volume one ends with the twelfth century ce. The work may be usefully
compared with

R. Walzer, chapter on 'Islamic Philosophy' in History of Philosophy Eastern
and Western, ed. Sarvapelli Radhakrishnan, Vol. II, pp. 120-148 (George
Allen and Unwin, London, 1953). Close-packed, judicious, illuminating
introduction to selected aspects.

Muhsin Mahdi, ed. {with R. Lerner), Medieval Political Philosophy (Free
Press, Glencoe, 111., 1963). The selections from Muslim and Jewish philo-
sophers are more enlightening on general philosophy than the title would
suggest, but omit the later Middle Period philosophers. (The selection of
Muslim writers does not fully correspond with that of Christian writers.)

Aldo Mieli, La Science arabe et son role dans revolution scientifique mondiale
(E. J. Brill, Leiden, 1938). Concentrates on that portion which led into
West-European Medieval science, but is quite comprehensive in scope ; it
is an alternate to the relevant sections in George Sarton's monumental
multi- volume Introduction to the History of Science.

Paul Kraus, Jdbir ibn Hayydn, Contribution a I' histoire des idees scientifiques
dans I'Islam, in Memoires de I'Institut d'Egypte (Vol. XLIV, 1933, and
XLV, 1942); a study of the most important early Islamicate chemical
corpus.

Fazlur-Rahman, Prophecy in Islam (George Allen and Unwin, London,
1958). A keen analysis of the Faylasufs' interpretation of Prophecy.

A. J. Wensinck, The Muslim Creed, Its Genesis and Historical Development
(Cambridge University Press, 1932). A study of the development of
Muslim theology through the earlier periods ; to be read along with

510 A SELECTIVE BIBLIOGRAPHY

W. Montgomery Watt, Free Will and Predestination in Early Islam (Luzac,
London, 1948). Probes into the motive forces in Muslim theological
development, but is not so comprehensive as Wensinck.

Louis Gardet and M.M. Anawati, Introduction a la theologie musulmane, essai
de theologie comparee (Vrin, Paris, 1948). A profound Catholic analysis,
restricted chiefly to the earlier period, and to some later Arabs.

W. Montgomery Watt, Islamic Philosophy and Theology (Edinburgh Uni-
versity Press, 1962), 'Islamic Surveys No. 1'. A brief introduction for the
beginning student.

Richard J. McCarthy, The Theology of al-Ash'ari (Imprimerie Catholique,
Beirut, 1953). Translations of two texts, notes, bibliographies; basic
material on this crucial theologian.

On Islamicate intellectual attitudes in the Middle Periods (supplementing
section on philosophy above) :

Soheil M. Afnan, Avicenna, His Life and Works (George Allen and Unwin,
London, 1958). A competent study, with bibliography of the many works
which have made Avicenna much better known than other Faylasufs;
note especially the studies and translations of Miss A.-M. Goichon.

Juveyni, Imam al-Haramayn, al-Irshdd (el-Irchad), ed. and translated into
French by J.-D. Luciani (Ernest Leroux, Paris, 1938), a complete and
detailed treatise of kalam by Ghazali's master.

Wm. Montgomery Watt, The Faith and Practice of al-Ghazdli (George Allen
and Unwin, London, 1953), translation (with good short introduction) of
two treatises by Ghazali, the first being his so-called spiritual autobiography
(Munqidh min al-daldl), the second an introduction to the practices of
the devout life.

G.-H. Bousquet, Ghazali: Vivification des sciences de lafoi (Iky a 'ulum al-din)
(M. Besson, Paris, 1955). A summary of Ghazali's great masterpiece.

A. J. Wensinck, La Pensee de Ghazzdli (Adrien-Maisonneuve, Paris, 1940),
a study of Ghazali's spiritual position, with special emphasis on his
Sufism ; perhaps it underestimates what was genuinely Muslim in the best
of Ghazali.

Averroes [Ibn-Rushd], Tahdfut al-tahdfut (The Incoherence of the Incoherence),
translated by Simon van den Bergh, 2 vols. (Luzac, London, 1954), an
excellent translation of Ibn-Rushd's response to Ghazali, which also
includes much of Ghazali's own work.

Henry Corbin, Suhrawardi d'Alep (d. 1191) fondateur de la doctrine illumina-
tive (ishrdqi) (Publications de la Societe des etudes iraniennes, No. 16,
Maisonneuve, Paris, 1939), a pamphlet by the man who has done more
than any other, in his subsequent work, to illuminate the later Sufi

A SELECTIVE BIBLIOGRAPHY 5H

thought of Iran; notably I' Imagination creatice dans le soufisme d'lbn-
'Arabi (Flammarion, Paris, 1958) and the most easily read Terre celeste et
corps de resurrection, de I'Iran Mazdeen a I'Iran Shi'ite (Buchet/Chastel,
Correa, i960).

Muhyi-d-din Ibn-'Arabi, La Sagesse des Prophetes [Fugue al-Hikam),
translated by Titus Burckhardt (Albin Michel, Paris, 1955), an abridged
translation of one of the two most important works of Ibn-al-'Arabi.

A. E. Affim, The Mystical Philosophy of Muhyid-Din-ibnul- Arabi
(Cambridge University Press, 1939), a metaphysical study of Ibn-al-
'Arabi.

Reynold Nicholson, The Idea of Personality in Sufism (Cambridge University
Press, 1923; repr. Muhammad Ashraf, Lahore, 1964). A careful study
especially of later Sufi thinkers on this point.

Abii-Bakr Ibn-Tufayl, The History of Hayy ibn-Yaqzan, translated from
Arabic by Simon Ockley and A. S. Fulton (Stokes, New York, 1929). A
rendering of Ibn-Tufayl's main work, minus its important preface.

H. S. Nyberg, Kleinere Schriften des Ibn-al- Arabl (E. J. Brill, Leyden, 1919),
a masterpiece of scholarly elucidation.

Muhsin Mahdi, Ibn Khaldun's Philosophy of History (George Allen and
Unwin, London, 1957). The best introduction to this important
philosopher.

Glossary of Selected Terms and Names

Listings in the Glossary are technical terms frequently appearing in the
text. Definitions and explanations given in the Introduction or elsewhere
of other terms, including geographical designations, may be located by
consulting the Index.

adab: the polite, literary culture of a cultivated individual; in modern
Arabic, 'literature'. A cultivated individual is an adib (pi. udabd').

Ahl al-Hadith: see Hadith folk.

ahl al-kitab: people of the Book, the Bible; that is, Jews and Christians,
though other groups sometimes were included.

'Alid: a descendant of 'Alt, cousin and son-in-law of the Prophet; the Shi'is
believed certain 'Alids should be imams (q.v.). 'All's first wife was
Fatimah, the Prophet's daughter, 'Ali's descendants by her (the only
descendants of the Prophet) were in particular called Fatimids. Des-
cendants of her son Hasan are often called sharifs: those of her son
Husayn are often called sayyids.

'alim (pi. 'ulamd'): a learned man, in particular one learned in Islamic legal
and religious studies.

Allah: an Arabic (both Muslim and Christian) name for the One God.

amir (also emir): a general or other military commander; after classical
'Abbasi times many independent rulers held this title; sometimes assigned
to members of the ruler's family. Amir al-mu'minin, commander of the
faithful, was the proper title of the caliph; amir al-umard' meant supreme
commander, generalissimo: used especially of the military ruler during the
decline of the High Caliphate.

Ansar: 'helpers' of Muhammad at Medina; name given collectively to the
Muslims native to Medina in distinction from the Muhdjirun (q.v.), those
who came with Muhammad from Mecca.

'aql: 'reason', 'reasoning'; in Islamic law systematic reasoning not limited
to qiyds (q.v.).

batin: the inner, hidden, or esoteric meaning of a text; hence Batinis,
Batiniyyah, the groups associated with such ideas. Most of these groups
were Shi'is, particularly Isma'ilis.

Dar al-Islam: lands under Muslim rule; later, any lands in which Muslim
institutions were maintained, whether or not under Muslim rule. It is
converse of Dar al-Harb, i.e., lands under non-Muslim rule.

dhimmi (also zimmi): a 'protected subject', follower of a religion tolerated

513

514 GLOSSARY

by Islam, within Muslim ruled territory, cf. ahl al-kitdb. The protection

is called dhimmah.
diwan (also divan): a public financial register; or a government bureau, or

council; or its chief officer; also the collected works of a poet.
Falsafah: philosophy, including natural and moral science, as expounded,

on the basis of the Greek tradition, in the Islamicate society. A Faylasdf

(pi. Faldsifah) is an exponent of Falsafah.
fard 'ayn: religious duty such as ritual worship, fasting, etc., incumbent

upon every individual Muslim excepting only those incapacitated from so

doing; fard kifayyah is collective duty such as jihad (q.v.) incumbent upon

the Muslim community as a whole, but so long as some are fulfilling it,

others may be excused,
fatwa: the decision of a mufti (q.v.).

fiqh: jurisprudence; the discipline of elucidating the Shari'ah (q.v.); also the
resultant body of rules. Afaqih (pl.fuqahd) is an exponent oifiqh.

ghazi: a warrior for the faith carrying out jihad (q.v.); sometimes applied
to organized bands of frontier raiders.

hadith (pi. ahddith; also hadis) : a report of a saying or action of the Prophet,
or such reports collectively. Sometimes this is translated 'tradition', as
having been transmitted from reporter to reporter; it has nothing to do
with tradition in the ordinary sense of anonymously inherited group lore.

Hadith folk (Ahl al-Hadith): people of the hadith, that is, those who con-
centrated on hadith reports as the chief form of learning and source of
authority; a particular type of piety was associated with them among
the Jama'i-Sunnis.

hajj: the annual pilgrimage to Mecca in the month of Dhu-1-Hijjah, the last
month of the Muslim year; required of every Muslim once in his life if
possible.

Hanaf i: referring to the Sunni legal madhhab (q.v.) ascribed to Abii-Hanifah

(699-767 CE).
Hanbali: referring to the Sunni legal madhhab (q.v.) ascribed to Ahmad

ibn-Hanbal (780-855 ce).

Hashimids: members of the family to which Muhammad belonged; its chief
branches were the 'Alids and the 'Abbasids. Hashimi usually refers to
that family, but may refer to the Shi'i group which regarded the imamate
as passing through Abu-Hashim, grandson of 'Ali.

hijrah: the flight of Muhammad from Mecca to Medina; the year it occurred,

622, is the base-year of the Muslim era.
hisbah: overseeing of public morals, including especially fair dealing in the

market; confided in High Caliphate to a muhtasib; later other terms were

also used for such an officer.

GLOSSARY 515

hiyal (sing, hilah): 'legal tricks', legal method of avoiding inappropriate
special consequences of general legal principles.

ijma': agreement of the Muslim community, as a ground for legal decision;
it was debated what constituted the community for this purpose.

'ilm: learned lore; particularly, religious knowledge, of hadith (q.v.) reports,
oifiqh (q.v.), etc. In modern Arabic the word is used to render 'science'.
Among many Shi'is it was supposed the imam (q.v.) had a special secret
knowledge, 'ilm.

imam: leader of the saldt (q.v.) worship; or the leader of the Muslim com-
munity. Among Shi'is: 'All and his descendants as proper leaders of the
Islamic community even when rejected by it, are held to have a spiritual
function as successors to Muhammad. Among Jama'i-Sunnis, any great
'dlim (q.v.), especially the founder of a legal madhhab (q.v.), was called
an imam.

iman: religious faith; conviction; that which a Muslim acknowledges both
inwardly and outwardly through his actions.

isnad (also sanad) : the series of transmitters of a hadith (q.v.) report, men-
tioned to guarantee its validity.

Jama'i-Sunnis: see Sunnis.

jihad: war in accordance with the Shari'ah (q.v.) against unbelievers; there
are different opinions as to the circumstances under which such war
becomes necessary. Also applied to a person's own struggle against his
baser impulses.

jizyah: commonest term for the 'poll-tax' paid by dhimmis (q.v.) in a Muslim-
ruled society, instead of the dues levied on Muslims.

kalam: discussion, on the basis of Muslim assumptions, of questions of
theology and cosmology; sometimes called 'scholastic theology',

katib: a secretary or scribe, specifically one who served in the government
bureaus.

Kharijis (also Khawarij; more properly Shurat): members of a group of
puritanical Muslim sects of Marwani and early 'Abbasi times.

khutbah: sermon delivered (by a khatib) at the special Friday midday
worship in the mosque; it contained prayers for the sovereign and, being
named in the khutbah and on coins, became a declaration of sovereignty.

madhhab (pi. madhdhib): a system oifiqh (q.v.), or generally the system
followed by any given religious group; particularly, four madhdhib were
ultimately accepted as legitimate by the Jama'i-Sunnis while Shi'is and
Kharijis had other madhdhib. Sometimes rendered 'sect', 'school', 'rite'.

Maliki: referring to the Jama'i-Sunni legal madhhab (q.v.) ascribed to
Malik b. Anas (715-795 ce).

516 GLOSSARY

mawla (pi. mawdli): master or servant, also a man of religious authority;
especially in the plural form mawdli it refers to persons associated with
Arab tribes otherwise than by birth, particularly in Marwani times;
non-Arab converts to Islam.

mihrab: the niche in the inside wall of a mosque (q.v.) showing the direction
in which prayer should be performed.

mosque (Ar. masjid): any place of worship for Muslims where the saldt
(q.v.) is performed in a group; a major one, where official Friday services
are held, is called j ami' .

mufti: an expert in the Shari'ah (q.v.) who gives public decisions in cases of
law and conscience.

Muhajirun: those who accompanied Muhammad from Mecca to Medina.

muhtasib: see hisbah.

Murji'is: name applied to a non-Shi'i school of thought among the Piety-
minded opposition of Marwani times.

Mu'tazilis: a group among the Piety-minded of Marwani times who developed
a lasting school of kaldm (q.v.) stressing human free responsibility and
divine justice.

nass: explicit designation (of a successor by his predecessor), particularly
relating to the Shi'i view of succession to the imamate; it thus comes to
confer upon the successor a power of knowledge and understanding that
no one else has.

pir: a Sufi master, able to lead disciples on the mystical way.

qiyas: the principle of deriving fresh juridical decisions, by way of
analogy, from decisions as given in hadith (q.v.) or Qur'an; commonly
listed as one of the four usul al-fiqh recognized among Jama/i-Sunms.

ra'y: personal judgment in working out jurisprudence rules.

salat (or saldh): the ritual worship performed five times daily, preferably
in groups; including the special group prayer of midday on Fridays. In
Persian called namdz.

sayyid: see 'Alid.

Shan't: referring to members of the Jama'i-Sunni legal madhhab (q.v.)
ascribed to al-Shafi'i (767-820 ce).

shahadah: the declaration that there is no god but God, and Muhammad
is His prophet, identifying the declarer as a Muslim.

Shari'ah (or Shar')\ the whole body of rules guiding the life of a Muslim,
in law, ethics, and etiquette; sometimes called Sacred Law (or Canon Law),
The provisions of the Shari'ah axe worked out through the discipline of
fiqh (q.v.), on the basis of usul al-fiqh (basic sources of legal authorities),
which Sunnis commonly list as Qur'an, hadith (q.v.), ijmd' (q.v.), and

GLOSSARY 517

qiyds (q.v.). Shi'is commonly substitute 'aql (q.v.) for qiyas and interpret
ijma' as consensus of the imams.

sharif: see 'Alid.

Shi' ah: 'party (of 'All)'; general name for that part of the Muslims that
held to the rights of 'Ali and his descendants to leadership in the com-
munity whether recognized by the majority or not; or any particular sect
holding this position. Shi'i is the adjective, or refers as a noun to an
adherent of the Shi'ah. Shi'ism (tashayyu') denotes the attitude or doctrines
of the Shi'ah. The most well known Shi'i groups are the Zaydis, the
Isma'ilis or Seveners, and the Twelvers.

shirk: associating other gods — or any objects of regard — with God; the
supreme sin.

Sufi: an exponent of Sufism (Ar. tasawwuf), the commonest term for that
aspect of Islam which is based on the mystical life.

sunnah: received custom, particularly that associated with Muhammad; it
is embodied in hadith (q.v.).

Sunnis: properly ahl al-sunnah wa-l-jama'ah ('people of custom and the
community'), in this work often Jama'i-Sunnis: that majority of Muslims
which accepts the authority of the whole first generation of Muslims and
the validity of the historical community, in contrast to the Kharijis and
the Shi'is (q.v.). Sunni as adjective refers to the doctrinal position, as
noun it refers to an adherent of the position. Sunnism is sometimes referred
to as 'Orthodoxy'. The term 'Sunni' is often restricted to particular
positions within the Jama'i-Sunni camp; e.g., often it excludes Mu'tazilis,
Karramis, and other groups which did not survive to command recognition.
In older Muslim works it sometimes included only the particular faction of
the writer.

taqiyyah: resort to pretence of conformity when avowal of one's principles
would be dangerous to oneself or others; used especially by Shi'is.

ulama.': see 'dlim.

Ummah: any people as followers of a particular prophet, in particular the
Muslims as forming a community following Muhammad.

'ushr: the tenth of the produce, a zakdt (q.v.) tax leviable on certain Muslim-
held lands as recognized by Islamic law.

usul al-fiqh: see Shari'ah.

zakat (or zakdh) : a fixed proportion of his property that is supposed to be
paid yearly by the Muslim of some wealth as a tax for the purposes of
a Muslim government, or else as charity to the poor or for other good
causes; rendered 'poor tax', or 'legal alms'.

Index

Note: Names and terms beginning with the Arabic definite article (al-) are
indexed under the letter following the article.

'Abbasid ('Abbasi) caliphate

administration under, 286-87, 292-94,

349-50, 493
bases of power and legitimation of

authority of, 284-85, 2gg, 348-50,

444-45
civil war (see Fitnah wars, Fourth)
court culture and patronage of art
and learning by, 283, 294-98, 303-4
(see also adab culture; adib)
crises and breakup of, 486-95 passim
and imperial absolutist ideal, 280-85

passim, 291, 444-45, 495
law under, 347-48 (see also Shari'ah)
Marwanis overthrown by, 273-75
revolts against, 313-14, 475-78, 482,

487-88, 490-92
and rise of local dynasties, 475-95

passim
urban prosperity and development of
new Islamicate culture under, 235-39
passim, 303-5
'Abd-al-'Aziz b. Marwan, 223, 258
'Abd-AUah, eldest son of Shi'i imdm

Ja'far, 376, 377
'Abdallah Ibn- 'Abbas, 254, 259, 273
'Abd-AUah Ibn-Mu'awiyah, Shi'i imdm,

273-74
'Abd-AUah Ibn-al-Zubayr. See Ibn-al-

Zubayr, 'Abd-AUah
'Abd-al-Malik, Marwani caliph, 221 n,
229, 243, 249, 284
administration of, 226, 243, 245-47
expeditions against Byzantine empire
by, 226, 268
'Abd-al-Rahman, Umayyad amir in

Spain, 309
'Abd-al-Rahman b. 'Awf, 169, 354, 356
'Abd-al-Rahman Ibn-Rustam, 310
'Abd-al-Wahhab, 64
Abraham, 128, 178, 179, 195, 253, 267
Abrahamic traditions, 128, 410. See also
Irano-Semitic cultural traditions
ideal of social justice in, 130-34

passim
and Islamicate culture, 447-49, 450-51
passim
absolutist ideal

in 'Abbasi regime, 280-81, 283-85,
291, 347-4 8 . 444-45. 495

Marwani move toward, 241-47, 271-72
in Sasanian empire, 140-41, 280-83,

284, 449
and Shari'ah-minded, 280, 299, 347-50
passim
Abu-Bakr (b. Abi-Quhafah), first caliph,
167-69, 172, 180, 198, 200, 207, 214
Abu-Dharr, 248

Abu-Hanifah, 255, 264, 318, 321, 335, 339
Abu-Hashim, 273
Abu-Lahab, 170, 171
Abu-Musa al-Ash'ari, 215
Abu-Muslim, 273-74, 276, 313-14
Abu-Nasr al-Fa.ra.bi. See Fa.ra.bi, Abu-
Nasr al-
Abu-Nuwas, 294, 297, 462-63
Abu-Salamah, 275
Abil-Sufyan, 190, 194, 213
Abii-Talib, uncle of the Prophet, 167,

170-71, 273
Abu-Tammam, poet, 464
Abu-'Ubaydah, philologian, 461
Abu-Yazid (Bayazid) al-Bistami. See

Bistami, Bayazid (Abu-Yazid) al-
Abu-Yusuf, prominent legist of Iraq, 62,

294. 335. 37°
Abu-Yusuf Ya'qub al-Kindi. See Kindi,

Abu-Yusuf Ya'qub al-
Abu-1-' Abbas al-Saffah. See Saffah

(Abu-1-' Abbas) al-
Abu-l-'Atahiyah, poet, 297
Abii-l-Hudhayl, 438
Abu-1- Khattab, Shi'i theorist, 370
Abu-I-Saraya, Sht'l rebel, 475
Abyssinia, Abyssinians

Arab naval expedition against, 204
conquest of Yemen and invasion of

Arabia, 151, 156
Muslim migration to, 169, 170-71
Academy of Athens, 315
Achaemenid empire, 118, 119, 138, 140,

282
adab culture

central position of Arabic poetry and

literature in, 452-53, 457~ 6 9
defined, 239, 445
under Samanids, 493
Sasanian and biblical traditions and,

448-51, 452, 454-55. 461. 465
and visual arts, 469-72

519

520

INDEX

adib, 291, 296
cultivation of: Arabic prose and poetry,
452-53. 457-69; history, geography,
and sciences, 453-57; interest in
arts, 469-72
'ulamd' and, differences in morality,
451-52, 461, 473-74
administration

'Abbasid, 286-87, 2 9 2_ 94> 349 _ 5°. 493
Marwant, 241-47 passim, 251, 270-72
under primitive caliphate, 208-13

passim, 218
Samanid, 493
Afro-Eurasian, 31, 91, 112
Afro-Eurasian Oikoumene, no, 114, 125

trade in, 124, 142
Afshtn, al-, 'Abbasid general, 482-83
Aghlabid dynasty, in Maghrib, 312, 475,

492
Agrarian Age, 53
denned, no
land most dependable source of income

in, 301
practice of medicine in, 419
agrarianate citied society, 50, 107-9,

109 n
agriculture

under 'Abbasids, 286, 295, 483-85
in agrarianate citied society, 107-9
in Axial Age, 121-24 passim
under Marwanis, 223, 245, 286
Muslim contribution to, 415
under Sasanian empire, 132-33, 142,
201-3, 212
AM al-Bayt (people of the house) , 260
AM al-HadUh. See Hadith folk
Ahmad b, Abl-Du'ad, 480
Ahmad b. Hanbal, 335, 352, 389-92

passim, 480
Ahmad b. Tiilun, 487, 488, 490
Ahura Mazdah, 128, 132, 141, 307, 316
'A'ishah, 180, 183, 184, 195, 2I 4
Akhtal, al-, 229-30
Akkadians, no
alchemy, 416-18, 423
alcoran. See Qur'an
Aleppo, Syria, 433, 464, 494
Alexander the Great, 119, 140, 410
Alexandria, 298
algorism, 414
'Ali b. Abi-Talib

acclaimed caliph, 214
civil war and arbitration, 214-16
figure of, as rallying point for the
Shi'is, 83, 216-17, 222, 251, 258-67
passim, 372-84 passim
and Kharijis, 216
loyalty to Muhammad of, 167

opposition to 'Umar and Uthman's
policies by, 213
'Ali al-Mas'udi. See Mas'udi, 'AH al-
'A1I al-Rida, 376, 479-80
'Alid-loyalism

defined, 260, 372

of Isma'ilis, 380

and sectarian Shi'ism, 278-79, 394

of Zaydis, 372
'Alids

and 'Abbasids, 275, 285, 291, 300

and caliphate, 259-66 passim

in Maghrib, 310-12

in Zaydl doctrine, 372
Allah, 160, 162, 170, 172, 195, 307

concept of, defined, 63, 73, 73 n

Creator-god of pagan Arabia, 155-56,
T 59
Allat, goddess of pagan Mecca, 156
Altheim, Franz, 44
Amin, al-, 'Abbasid caliph, 299-300
amir ('commander'), title of Umayyad

ruler in Spain, 309
amir al-mu'minin ('commander of the
faithful'), title taken by Caliph
'Umar, 207
amir al-umard' (commander-in-chief), 494,

495
Amorium, 203, 482
Amr (Command), 166
'Amr, Saffarid governor, 487
And'l-haqq ('I am the Truth'), 405
Anatolia, 112, 226, 140

Muslim expeditions against, 212, 219,

268, 482
Ankara, 482
Ansdr ('helpers'), 174, 176, 197-98,

254-55
Antioch, 298, 494
'Aqabah, 171

'aql (reasoning), in Shari'ah law, 329
'Arab (Arab): usage and meaning of,

62-63. r 5°
arabesque, 75
Arabia. See also Bedouin
under 'Abbasids, 314
pre-Islamic: camel-nomadism, 147-50;

culture, 153-54, 196, 227-30 passim;

religions, 155-57 passim; trade and

trade routes, 150-54 passim
Arabic language, 3, 40—43 passim, 151-53

passim
and adibs, 452
as common language of culture in the

High Caliphal Period, 314
development of technical vocabulary

in, for scientific studies, 413, 413 n

INDEX

521

Arabic language — contd.

and grammar, 238, 254, 297, 357, 452,

468-69
influence on Pahlavi of, and beginning

of New Persian, 450-51
script of, 75, 247
spread of, in Fertile Crescent and

Egypt, 449
translations into: from Greek, 235, 239,
298, 412, 429-30; from Pahlavi, 296,
298, 412; from Sanskrit, 298, 412
transliteration of {see transliteration),
usage of, in administration and
commerce in Nile-Oxus region,
235-36, 246, 251
Arabic literature

central position of, in 'Abbasid court

culture, 296, 452-53. 4 6 5- 6 9
poetry, I53~54. l8o > 229-3°. 235-36,
239, 297. 457- 6 4
Arabism, 320, 325, 448
Arabs. See also Bedouin; Quraysh
client-kingdoms of, 150, 153-54
conquests of, 198-206 passim, 212, 227
development of camel-nomadism

among, 147-50
neo-tribalism and factional rivalries

among, 227-29 passim, 249, 252
relations of, with Sasanian and Roman
empires, 144, 151-54 passim, 156,
198-99
settlement of, in conquered territories,
203-4 ( see a ^ so garrison towns)
Aramaeans, 196

Aramaic, 137, 140, 201, 203, 410, 449
in Axial Age, m-15 passim, 118-19
language of Arab merchant-kings, 153
Aristotle, 239, 425, 430, 451

system of astronomy of, 413-14, 426,

427
Armenians, 306
armies

of 'Abbasids, 481-82, 485-87, 493-94
passim

of Marwanis, 273

of primitive caliphate, 212, 217-19

of Sasanians, 143-44
Arsacid dynasty, 119, 137
art and architecture, 246-47

in High Caliphal Period, 469-72 passim

iconophobia in, 247, 306, 368-69
asceticism, 238, 394, 400
Ash'ari, Abu-1-Hasan al-, 439-40, 442
Ash'aris 65-66, 439-43
Asma'i, al-, grammarian, 459
associates of Muhammad, 214, 214 n,

254-55. 276, 327
Assyrian empire, no, 118

astrology, in, 239, 419, 420, 421
astronomy

Hellenistic tradition of, 411, 413,

426-27
Islamicate studies in, 414-15, 416,

421-23 passim, 425
translation of works on, into Arabic,
412, 414
Athens, 144

atlas, 55-56, 455-56 passim
Attica, 33

Avicenna. See Ibn-Sina
'Aws, tribe of Medina, 172
Awza'i, al-, foremost legist of Syria, 321,

335. 37°
Axial Age, 108, 118, 121, 125, 142, 446,

45°

defined, 112, 114 n

traditions of, m-17 passim
Azerbaijan, 314, 475, 482
Azraqis, 221

Babak, 314, 482

Babylon, Babylonians, 103, 115, 119,
138, 421

Badr, battle of, 176, 179, 180, 194

Baghdad

'Abbasid court at, 294-98 passim
under 'Abbasids, 280, 283, 290-91,

292-95 passim, 475, 480, 482, 488
Buyid occupation of, 495
as cultural and economic centre of

High Caliphate, 287, 298
foundation of, as capital, 275, 280, 287
science and philosophy in, 298
shift of power from, 488, 493
siege of, 300

Bahrayn, 194, 491

Bakht-Ishu', Nestorian physician, 298

Baladhuri, al-, 455

Balkan peninsula, 138

Balkh, 295, 456

Banu Ghassan, 195

Banu Hanifah, 197

Banu Hashim, 158, 170, 171

Banu Nadir, 190, 191

Banu Qaynuqa.', 177, 178, 190

Banu Qurayzah, I9r

Banu Umayyah, 213, 214, 219. See also
Umayyad caliphate

barid. See postal service

Barmakid family, 294-95

Barmecides. See Barmakid family

Basetti-Sani, Giulio, 29

Basrah, 208, 214, 222, 245, 258, 271, 276

bdtin ('inward meaning'), 266, 379, 473

Batiniyyah (Batini), 238-39, 379. 380,
383, 393-94

522

INDEX

Bayazid al-Bistami. See Bistami, Bayazid

(Abu-Yazid) al-
Baykand, 227
Bayt al-Hikmah ('House of Wisdom'),

in Baghdad, 298
Becker, Carl, 28, 31
Bedouin, 32, 41, 62, 197

alliance with Muslims of, 192-93, 194
as allies of Quraysh, 190-92, 194
blood feuds among, 149, 155, 167
camel-nomadism among, 147-50
code of honour of, 149, 167, 170
conquest of Yemen by, 204
control of trade routes by, 152—53
Muslim campaigns against, 198
poetry of, 153-54, J 8o, IO A 229-30,

239, 297, 458-64 passim
social and economic organization of,
148-51
belief: Mu'tazili doctrine of, 385, 439
Berbers, 206, 219, 226, 269, 308-12

passim
Bethmann, Eric, 29
bid'ah ('innovation'), 252, 321, 324, 325,

357. 438
Biruni, al-, 421
Bistami, Bayazid (Abu-Yazid) al-, 398,

404-5, 405 n, 408, 409
Black Stone, 156, 195, 491
blood feuds. See Bedouin, blood feuds

among
Book, The, 297
Book of Misers, 466
Bridge to Islam, 29
Buddha, 125
Buddhism, Buddhists, 125, 128, 143, 144,

226, 238
Bukhara, 227

Bukhari, al-, compiler of hadiths, 332, 473
bureaucracy. See administration
Buyid (Buwayhid) dynasty, 494-95
Buzurgmihr, Sasanian vizier, 282
Byzantine empire, 191, 192, 204, 314.
See also Roman empire
Arab raids on, 198-99
expedition of, against Hamdanids, 494
Muslim expeditions against, 212, 219,

226, 268, 482
usage of term defined, 54
Byzantium, 33, 156, 195, 246, 247

Cahen, Claude, 41

Cairo, 302

calendar, Islamicate, 20-22, 211, 307

caliph

origin of the institution and basis of
authority of, in primitive caliphal
state, 86, 198, 207-8, 214, 217-18

principle of dynastic succession of, 223,

247, 299-30, 348-50
role of, in 'Abbasi and Marwani states,
223, 246-47 passim, 280-81, 283-85,
291-94 passim, 299-300, 348-50
caliphate: and 'ulama", 299—300, 348—50,
473-74- See also High Caliphate;
primitive caliphate; and under
individual caliphal states
calligraphy, 471-72
camel-nomadism, 147-50
Carmathians. See Qarmatians
Carthage, 312
Caspian Sea, 268, 310
ceramic art, 471
Chalcedon, Council of, 201
Chalcedonian Church, 140
Charlemagne, 310
China, 112, 125, 142, 233, 237
Ch'in-Han empire, 119
Christianity, Christians. See also dhimmis
under caliphates, 199, 206, 227, 229,

235-36, 250, 305-8 passim
in Iraq, 203

in Maghrib and Spain, 309-10
in Nile-Oxus region, 157
as participants in Muslim culture, 90
personal piety of, 305-8 passim, 361-62
and Socratic philosophic traditions,
138, 412
Cilicia, 203, 494
circumcision, 324
citied society. See agrarianate citied

society
cities

under 'Abbasids, 287, 295-96, 301-5
passim, 347-48 (see also Baghdad)
in ancient agrarian-based society,

105-6
in Maghrib, 308-12 passim
civil wars. See Fitnah wars
civilization
denned, 30-34

history of studies on Islamicate, 39-45
method in historical studies of, 22-26
philological and Arabistic emphasis in
studies of Islamicate, 31-32, 39-43
scholarly precommitment and study

of Islamicate, 26-30
terminology of studies on (see
terminology, scholarly)
coinage, 226, 246—47
commercial law. Islamic, 338-39
community. Islamic. See Jamd'ah;

Ummah
Companions of the Prophet. See associates
of Muhammad

INDEX

523

Comparative Studies in Society and

History, 41
confessional religious traditions
rise of, in the Afro-Eurasian

Oikoumene, 125-28
tendency of, to reform all social
patterns, 315
Confucianism, 315
Constantinople, 137, 204, 206

besieged by Muslims, 191, 219, 226,
268, 268 n
contracts and obligations, Muslim law of,

338-39
conversion to Islam

under 'Abbasids, 301-8 passim

of Aramaeans and Persians, 230

of Bedouin, 192, 194

of Berbers, 226, 308

under Marwanls, 226, 269-71, 305

and Piety-minded, 252-54
Coptic, 140, 203
C6rdova, 310
cosmology

of Ash'arls, 440-42

of Faylasufs, 424-26, 427-28, 437,
440-42 passim

of Isma'ilis, 431

of Mu'tazilis, 437-38

neo-Platonist, 426-27

of al-Razi, 430
craftsmen, 469—70
Cragg, Kenneth, 29
Crusaders, 206

Ctesiphon, 138, 201, 206, 208, 287
cult, Muslim, 72-78 passim, 81, 209-11,

368
cultural traditions. See Irano-Semitic

cultural traditions
Cuneiform no 112, 115, 119, 410
Cyprus, 212

dd'is, Isma'ili missionaries, 380, 490, 492

Dalmatia, 138

Damascus, 200, 208, 229, 243, 300

Ddr al-Isldm, 343

Darius, 282

David, 450

Daylam, Daylamis, 494, 495

denarius, Byzantine gold coin, 246

devotional tradition: Muslim, 363-66

passim, 371
dhimmts ('protected subjects'), 229,
2 42-437 245

under caliphates, 302, 304-8 passim,
322

and conversion, 269, 305

and Kharijis, 258

piety of, 305-8 passim

and Piety-minded, 447-48, 449
and taxes, 269-70
and trade, 302, 305
diacritical marks, in transliterating

Islamicate languages, 5 n, 10 n. See

also transliteration
dinar, gold coin, 246
divine will: direct human responsibility

before, in Islam, 318-20
diwdn, public financial register in

primitive caliphate, 208, 211, 270
diwdn (bureau), under 'Abbasids, 292
Diwdn (collected verses), of Imru'-al-

Qays, 459
Diyalah river, in Iraq, 483
Dome of the Rock, 247

Eastern Roman empire. See Roman

empire; Byzantine empire
egalitarianism

in Irano-Semitic monotheism, 134-35

passim
in Islam: challenged by ideal of
absolutism, 241; before God, 18 1,
281, 318; of garrison towns, 343; of
Piety-minded, 253, 304, 317-18;
of Shar'i, 281, 340, 344
Egypt
under 'Abbasids, 274, 300
Arab conquest of, 203-6
under Marwanis, 222, 223, 258, 269
under Roman empire, 120, 136, 137,

138
under Tuliinids, 487, 490
Ein asiatischer Staat: Feudalismus unter
den Sasaniden und ihren Nackbam,

44
Encyclopaedia of Islam

Arabistic and philological bias in, 40
transliteration of Islamicate languages,
system of, 8 n, 9-10, 10 n, 11, 13,
15-16
Epistles of the Pure Brethren [Ikhwdn

al-Safd'), 384
era, Islamic. See Hijrah
esotericism

among Jama'i-Sunni, 393-94
among Shi'is, 374, 379, 383, 393-94
Euphrates, 201, 214, 274

Fadl b. Yahya, 295
'Faith', Ash'ari's doctrine of, 439-40
Falsafah fphilosophia'), 419-37 passim,
440

and adibs, 451, 452

defined, 239, 418
family law, 181-82, 211, 340-43
faqih (jurisprudent), 335

524

INDEX

Farabl, Abu-Nasr al-, 433-36

Farazdaq, al-, 230

fari'ayn, obligatory duties, 322

fard kifdyah, community duties, 322

Fars, 140, 212, 488

Fas (Fez), Morocco, 310

Fdtihah, opening surah of Qur'an, 185

Fatimah, daughter of Muhammad, 219,

259, 260, 378
Fatimid dynasty, 492, 495
fatwa, decision on a point of Shari'ah

by a mufti, 349
fay', state property, 208, 213, 214
Faylas&f ('philosopher'), 418-37 passim
and 'Abbasid court, 419, 437, 444-45,

473-74

and Ash'ari haldm, 440-41

defined, 418

Hellenic traditions and, 448-49

and medicine, 419-22

and Qur'anic traditions, 428-37 passim

study of astrology by, 421-22

'ulamd' and, 419
Fertile Crescent, 203, 239

in Axial Age, 115, 119

in Cuneiform times, 97, no, in

populist movements in, 143

trade routes of, 123, 130
Fez. See Fas
fiqh, Islamic jurisprudence, 255, 333-36,

453
firqah ('school of thought'), meaning of,

66-67
Fitnah wars

First, 214-17, 354

Second, 219-23, 226, 259

Third, 273-75 passim, 283, 300

Fourth, 300-301, 475-81 passim
Freud, Sigmund, 396-97, 397 n, 407
Fustat (Old Cairo), 208, 213

Gabriel, 162
Galen, 239, 419

garrison towns (amsdr), 208-11, 212, 213,
216

Arab tribal system reconstituted in,
227-30, 287, 325-26

and Kharijis, 222

in Khurasan, 219

non-Arabs in, 243, 249

shift of power to, 217
Gathas, 141
Gaul, 227, 309-10
Geber. See Jabir b. Hayyan
geography, 455-57, 457 n
geometry, 423
Ghassanids, 153
ghdzts, warriors for the faith, 494

Ghulat, Shi'i theorists, 265, 266, 370,

374. 379

Gibb, H. A. R., 68

Gilgamesh, 446

Gnostic, Gnosticism, 136, 143, 159

God

al- Ash'arl on, 439-43 passim
Faylasiifs' notion of, 425-28
Hadith folk on, 387-89, 391-92
monotheists' conception of, 428
in Qur'an, 262-64, 367-68
Shari'ah-minded emphasis on direct
responsibility before, 318-20, 336
in Sufi mysticism, 394-95, 398-402,
405, 408-9

Goldziher, Ignaz, 28, 65

Greek

church, 157, 201, 203

culture, 124

language, 136, 137, 138, 142, 410

philosophers attracted by Sasanian

school at Jundaysabur, 144
traditions, 103, 119, 239 (see also
Hellenic traditions)

Grunebaum, Gustave von, 31, 41

Gujarat, India, 302

Gurganj, Khwarazm, 295

Hadt, al-, 'Abbasid caliph, 291
hadith ('report')

definition and usage of, 63-66, 254,

254 n
development of, 326-27, 332
and Hadith folk, 386-87
ritual obligations determined on basis

of, 337
and al-Shafi'i, 328-33
Shi'i view of, 326
and al-Tabari, 351-57 passim

Hadith folk (AM al-Hadtth)

doctrine of, 385-89, 391-92, 439-40
al-Ma'mun's repression of, 389-90,

480-81
and al- Mutawakkil, 486
Mu'tazili condemnation of, 438
strength among Baghdad poor of, 391
Sufi association with, 393, 402

bajj (pilgrimage), 155, 162, 179, 195, 211,
292, 337

Hajjaj b. Yusuf, al-, 223-26, 245, 248,
249, 267

hakim (judge-arbiter), 173

hdl, Suit term, 406

Halakha, Jewish law, 317

Hallaj, al-, 405, 409

Hamdanids, 464, 494-95

Hamzah, uncle of Muhammad, 190

INDEX

525

Hanafis, madhhab of jurists, 333, 335,

343. 346, 387. 44°. 463
Hanbalis, madhhab of jurists, 65, 392
Hanifs, monotheistic Arabs neither

Christians nor Jews, r6o
Haqq ('Truth'), Sufi term for God, 401
Harran, 235, 298, 412
Harun al-RashSd, 'Abbasid caliph,

291-300 passim, 475
Harura', 215

Hasan, grandson of Muhammad, 216, 372
Hasan al-'Askari, 377
Hasan al-Basri, 248-49, 264, 386, 394
Hashim, House of (Hashimids), 259, 260,

273. 274
Hashwiyyah. See Hadith folk
Hebrew

Bible, 170, 198, 448
prophets, 72, 91, 115, 1 17-18
traditions, 103, ng, 120-21 (see also
Abrahamic traditions)
Hellenic traditions
and Faylas&fs, 448-49, 451
influence of, on Nile to Oxus region,

119-20, 137-38, 140, 410
revival of, by Nushlrvan, 144
revivification of, under the High
Caliphate, 235
Heraclius, 192, 201
Herodotus, 25
High Caliphate
defined, 233

evolution of new cultural patterns
under, 235-39, 303-5 (see also adab
culture)
Hijaz, 157, 161, 187, 199, 212, 219, 222,

248, 267, 291, 294
Hijaz-Syria axis, 245-46
Hijrah, 20, 21, 172, 211, 260
Hind, 190
Hindus, 90, 226
Hira,', Mount, 158, 160
Hirah, 153, 199, 203, 208
Hisham, Marwanl caliph, 271-72, 275,

305
Hisham b. al-Hakam, 437
Hittites, no
hiyal ('tricks'), 339, 346
Homer, 137

Hudaybiyah, treaty of, 193, 194
Hujrb. 'Adt, 218, 284
Hujwlri, 400
Hunayn b. Ishaq, 412
Hurgronje, Christiaan Snouck, 28
Hurrians, no
Husayn, grandson of Muhammad, 219,

222, 260, 372, 378

Ibadis, 335, 371, 492. See also Kharijis
Iblis (Satan), 165
Ibn-al-Ash'ath, 245
Ibn-al-Hanafiyyah, 222, 262, 265, 272,

273. 276, 376
Ibn-al-Muqaffa', 284-85, 296, 413, 466,

468
Ibn-al-Mu'tazz, 463
Ibn-al-Zubayr, 'Abd-Allah, 219, 221, 222,

222 n, 223, 258, 265
Ibn-'Amir, 212-13
Ibn-Durayd, 468
Ibn-Karram, 393
Ibn-Khaldvin, 62
Ibn-Mas'vid, 213
Ibn-Qutaybah, 461, 463
Ibn-Sina, 425
Ibn-Ubayy, 178
Ibn-'Umar, 259, 386
Ibrahim, 'Abbasid imam, 273, 274
Ibrahim, pretender to 'Abbasid throne,

475
Ibrahim al-Mawsili, 294
Ibrahim b. al-Aghlab, 312
Idris II, Idrisid ruler in Maghrib, 310
Idris b. 'Abd-Allah, 'Alid leader in

Maghrib, 310
Idrisid dynasty, 492
ijmd" ('consensus'), agreement of Muslim

community as ground for legal

decisions, 64, 324, 330, 333, 354
Ikhshid, 495
Ikhwdn al-Safa". See Epistles of the Pure

Brethren
'Urn ('knowledge'), 255-58
of Abdallah Ibn-'Abbas, 274-75
and 'Alids, 259-66 passim
and Piety-minded, 325
Shl'l doctrine of, 335, 345, 372-76

passim
'Urn al-rijdl ('knowledge of the men'),

332, 35°
im&m, imamate, 256-67 passim
and 'Alids, 259-67
and Shl'l groups, 326, 335, 373~76

passim (see also Isma'ills; Twelver

Shi'is)
Imamls. See Twelver Shi'is
imdn ('faith'), 174, 439
Imru'-al-qays, 459-60, 462, 463
Index Islamicus, 41
India, 74, 125, 130, 142, 144, 237
Indie, 49, 117, 140, 233, 237

influence on Sasanians, 144-45, 413
Indo-Persian languages, transliteration

of (see transliteration)
Indus valley, 226
inheritance. Islamic law of, 343

526

INDEX

Iran, Iranians, 204, 219, 221, 241, 274,

378
revolt in, against 'Abbasi regime,

313-14
Iranian highlands, 206, 212
Iranianism

influence of, on adibs, 448, 449-50
Irano-Semitic cultural traditions. See
also absolutist ideal, in Sasanian
empire; merchants, mercantile
classes, and Irano-Semitic traditions
defined, 61-62, 117, 117 n
growth of confessional religions within,

125-28
and Islamicate civilization, 43, 83,
91-92, 103-4, 233-38 passim, 280-83,
316-18, 445-51
populism in, 130-44 passim
Iraq, the, lower Tigris-Euphrates valley,
48, 125, 138, 151, 199, 208, 222, 245.
See also Sawad
under 'Abbasids, 275, 487
agricultural investment by Sasanians

in, 144, 201-3
Arab conquest of, 201—6 passim
Piety-minded opposition in, 248-49
'Iraq 'Ajami, plateau area of western

Iran, 204, 259, 494
irrigation, 212, 245-46, 286, 295
Ishmaelism, 252

Islam dans le miroiv de I'Occident, 28
Islamdom, Islamicate, definition and

usage of, 57-60, 95
Islamic studies, history of. See civilization,

history of studies on Islamicate
Isma'il, 376, 378-79
Isma'il b. Nasr, Samanid ruler, 492
Isma'ilis, 335, 376, 473

and cyclical concept of history, 381-83
doctrine of, 378-84

and esoteric Shi'ism, 379, 382, 383, 393
Hellenic influence on, 379, 431
and 'hidden' imam, 380
and lore of taqiyyah, 381
in Maghrib, 491-92, 494
and Qarmatian rebellion, 490-92
isndd (also sanad), chain of transmitters
of hadlth, 327, 328-29, 331, 352, 353,

454-55
Israelism, 252
Isrd'iliyydt (narrative tradition attributed

to Jews), 317
Istakhri, al-, 456
Italy, 138, 312
Ithna. 'asharls. See Twelver Shi'is

Jabir b. Hayyan (Geber), 384, 417

Jacobite Christianity, 157
Ja'far, of Barmakid family, 295
Ja'far al-Sadiq, 260, 335
jdhiliyyah, 174
Jihiz, 'Amr al-, 463, 466-68
Jahm b. Safwan, 233
Jalaluddin Rumi, 401
Jamd'ah (Muslim community). See also
Ummah

ideal of unity of, 248, 250, 258, 259

and Marwanis, 223, 245-46, 258, 265

and Mu'awiyah, 217-19

and Piety-minded, 276

Qadari support of, 264

and Shi'i-Sunni schism, 276-78
Jama'l-Sunni. See also Sunnls

and caliph, 336, 345, 348-50

and Shart'ah law, 326-35 passim
Jarir, poet, 230
Jaspers, Karl, 112
Jazirah, Mesopotamia, 201, 221-22, 245,

300
Jerusalem, 171, 178, 179, 191, 247, 316
Jesus Christ, 72, 165, 195

in Sufi mysticism, 398, 401-2
Jews, Judaism, 90, 128, 133, 135, 142-43,
201, 203. See also Abrahamic
traditions; dhimmts

in Arabia and Yemen, 156, 157, 172,
177-78

under caliphates, 227, 235, 305-8
passim, 309

and converts, 252-53

influence of, on Piety -minded, 316-18
jihad, 75, 269, 292
jinn (spirits), r62, 171
jizyah (poll-tax), 270
Jubba'i, al-, 442
Junayd, al-, Sufi master of Baghdad,

407-9 ^
Jundaysabur, 144, 298
jurisprudence. Seefiqh
justice

concept of, in 'Abbasid court, 444

concept of, in agrarian-based societies,
ro6

in Irano-Semitic traditions, 130-34
passim

Islamic ideal of equal, 369

Sasanian concept of, 140-42, 144, 282

in Shart'ah, 336-37, 344

and 'ulamd', 474
Justinian, 144, 315

Ka'bah, 155-56, 159, 169, 179, 195, 223,

267
Kaffir, 495
kdhin (soothsayer), 162

INDEX

527

kaldm ('discussion'), based on Muslim
assumptions concerning theology
and cosmology

Ash'ari school of, 439-43

definition and meaning of, 437-38

and Karramis, 393

Maturidi school of, 439-40

and Mu'tazilis, 385, 437-39
Kalb, Arab tribal bloc, 221, 229, 264,

272, 300
Kaltlak wa Dimnah, 296
Karbala,', 219, 222
Karramis, 393, 440
Karun (Dujal) valley, Khuzistan, 201
kdtib (scribe or secretary), 286, 295, 296,
299, 490

social origins of, 444-45
Khadijah, wife of Muhammad, 158, 160,

162, 167, 171, 180
Khalid al-Barmaki, 295
Khalid al-Qasri, 267, 272
Khalid b. al-Walid, 190, 198
khalifah. See Caliph
khardj (land-tax), 270
Kharijis (Shurat)

and 'Abbasids, 291, 300, 313-14

and 'Alt, 215-16

in Arabia and Iran, 221-22, 475, 487

egalitarianism of, 216, 221, 256

in Maghrib, 310, 313, 492

Marwani struggle against, 223, 245, 248,
272-73

and Shi'is, 262, 264, 371
Khazraj, Bedouin tribe of Medina, 172
Khidash, 'Abbasid chief in Khurasan, 273
Khurasan, 212, 216, 2ig, 226, 273

under 'Abbasids, 274-75, 2 $5> 2 95>
299-300, 305, 313, 475-82 passim
khuibah (sermon), 209
Khuzistan, 201, 298, 391
Khwarazm, 295, 391

Khwarizmi, Muhammad b. Musa al-, 414
Kindah, 153, 154, 459
Kindt, Abu-Yusuf Ya'qub al-, 430-31,

433
Kisa/i, al-, 294, 297
Kitdb al-Umm, 327
Kraemer, Jorg, 3 1
Kroeber, Alfred, 50
Kufah, 203, 212-14, 216, 218, 222, 245,

258-59, 264-65, 272-74 passim
kiific, style of Arabic script, 247
huff, 174
kunyah (honorific name), 17

Lakhmid, Arab kingdom, 153-54, IQ 9>

203, 208
landed gentry, 121, 132-33, 348-49

Last Day, Qur'an description of, 163
Last Judgment, 73, 130
Lewis, Bernard, 58
Light, Surah of, 184, 185
literary criticism, 357, 460-64
literature. See Arabic literature

MacDonald, Duncan, 28

madhhab ('chosen way'), system oifiqh

followed by any religious group, 335
Maghrib (North Africa), 138, 206, 219, 226,

308-12, 321 n, 492
Magi, hereditary priestly class of

Mazdeism, 141
Magianism. See Mazdeans, Mazdeism
Mahdi ('guided'), 276, 374
Mahdi, al-, 'Abbasid caliph, 289-91
Mahdi, al-, Fatimid caliph in Maghrib,

492
Mahdiyyah, Fatimid capital in Maghrib,

492, 495
mahr ('bride- wealth'), 182, 342
Maimonides, 59-60
Makdisi, George, 65-66
Malaysia, 17

Malaysian archipelago, 124
Malik b. Anas, 318-24 passim, 335, 387
Mdliki, madhhab of jurists, 312, 335, 346,

4 6 3
Ma'miin, al-, 'Abbasid caliph

builds new capital at Samarra, 472
and caliphal absolutism, 473 -
civil war and provincial rebellions

during reign of, 300, 475-79
and Hadith folk, 389-90, 480-81
interest in science and philosophy of,

298, 412, 414
and Mu'tazilis, 389, 478, 480-81
and 'ulamd' and adibs, 474
Manat, a goddess of pagan Mecca, 156
Mani, founder of Manicheanism, 290
Manicheanism, Manicheans, 125, 128,

136, 143. 379
foundation and doctrine of, 289-91
Mansur, al-, 'Abbasid caliph, 284-89, 296,

370, 386, 460, 466
maqdm (level), in Sufi mysticism, 406
Maqdm Ibrdhtm, sacred stone near the

Ka'bah, 179
maqsilrah, grille setting off part of mosque

for use by rulers or women, 292
marriage law

of Muslims, 181-83, 34 -43
of pre-Islamic Arabs, 181-82
ma'rif, al- ('the known'), 163
Marv, city in Khurasan, 300, 461
Marwan, counter-caliph in Syria, 221,

221 n, 222, 223

528

INDEX

Marwan II (b. Muhammad), Marwani

caliph, 272-74, 283
Marwani (Marwanid) caliphal state,
221 n, 226-27
basis of power of, 223, 229, 251
centralization of authority and

administration under, 241-47, 251,
270-72, 283-84
Piety-minded opposition to, 248-51,

253- 255, 259-75 passim
and principle of hereditary succession,
223, 247, 268
Marx, Karl, 105
Maslamah, 197
Massignon, Louis, 28, 29
Mas'udi, 'Alt al-, 455
mathematics, 119, 298, 410, 412, 414-15
Maturidi, Abu-Mansur al-, 439-40
Maturidi school of kaldm, 439-40
Maurya empire, 1 19
mawdli (sing, mawla; 'affiliates' or

'freedmen'), non-Arab converts to
Islam, 222, 229, 249-50, 265, 270,
273-74. 275
mazalim, courts of, 347-48
Mazdak, 45, 143

Mazdeans, Mazdeism, 128, 140, 144, 203,
206, 317-18
under caliphates, 227, 235, 307-8
ideal of social justice of, 132-34, 141
and Sasanian empire, 125, 141-43, 316
Mecca, Meccans, 223, 267, 310
and Muhammad, 158-72 passim
pilgrimage to, 155-56 (see also hajj)
position of, in pre-Islamic Arabia,

1 54-5 7
taken by Muslims, 194
medicine, 239, 298, 415-16, 419, 420, 421
Medina (Yathrib)

as centre of Muslim authority and

primitive caliphate, 198-217 passim
community at, as model for Shari'ah

legists, 320-26
'constitution' of, 173, 183
establishment of Muslim Ummah

(community) in, 172-193 passim
Jewish clans exiled from, 177, 190-91
under Marwanis, 247-48, 258, 268-69
Muhammad at, 172-94 passim
Mediterranean area, 137, 138, 142, 204,

235, 308-12 passim
Melitene, 203

merchants, mercantile classes, 107, 117,
124
in 'Abbasid state, 287, 301-5 passim,

390-91, 34 6 -4 8 . 452
and Irano-Semitic traditions, 130,

133-34. M2-43

Merida, Spain, 310

Middle East. See Nile to Oxus region

mihnah (inquisition), 389

mihrdb, niche in mosque indicating

direction of Mecca, 246
mi'rdj, 171

misr (pi. amsdr). See garrison towns
Modern, definition and usage of term,

50-52
monarchia! ideal, 281-83, 495
monasticism, 238, 317, 394
Monophysite Christians, 201, 203, 204,

306
Moses, 170, 185
mosque, 208-1 1, 305, 348
Mu'allaqat, collection of pre-Islamic

Arabic poetry, 459-60
Mu'awiyah, Umayyad caliph, 203, 212,

213, 214, 221, 229, 245, 269
reign of, 217-19
Mudar, bloc of Arab tribes, 229
Mudari Arabic. See Arabic language;

Arabic literature
Mufaddaliyydt, 460
mufti, 347
Muhdjiriln ('emigrants'), those who

accompanied Muhammad from

Mecca to Medina, 174-75, J 78. 214
Muhammad (Abu-1-Qasim Muhammad

b. 'Abd Allah), the Prophet
early life of, 158
emigration to Medina, 171-72
family life of, 179-80
lays foundation of new religious

community, 167-72
at Mecca, 158-72
at Medina, 172-94

personality and piety of, 167, 184-85
return to Mecca and death of, 194-95
Muhammad al-Baqir, 260
Muhammad al-Muntazar, 377
Muhammad al-Nafs al-Zakiyyah, 276, 284
Muhammad b. Abu-Bakr, son of first

caliph, 355
Muhammad b. Ishaq, 254-55
Muhammad et Saint Frangois, 29
Muhasibi, al-, 406, 480
Muhtadi, al-, 'Abbasid caliph, 486
muhtasib (market inspector and supervisor

of public morals), 347
Mu'izz-al-dawlah, Buyid dynasty, 495
Mukhtar b. Abi-'Ubayd, 222, 265, 266,

270, 276
mundfiqun ('waverers'), 178
Mu'nis, 'Abbasid general, 494
Muktaf !, al-, 'Abbasid caliph, 488, 492
Muntasir, al-, 'Abbasid caliph, 486
Muqanna', al-, 313

INDEX

529

muqdtilah (soldier-tribesmen), 212
Muqtadir, al-, 'Abbasid caliph, 493,

494-95
Murji'ls, 264-65, 266, 271
Musa al-Kazim, 262, 376
musddarah, 271, 283
Muslim, collector of hadttks, 332
Mu'tadid, al-, 'Abbasid caliph, 488
Mu'tah, 194

Mu'tamid, al-, titular 'Abbasid caliph, 487
Mutanabbi', al-, 464
Mu'tasim, al-, 'Abbasid caliph, 481-83
Mutawakkil, al-, 'Abbasid caliph, 392,

466, 485-86, 487, 491, 493
Mu'tazills, 394, 442

doctrine of, 384-86, 437-39

and Hadith folk, 388-89
Muwaffaq, al-, 'Abbasid caliph, 487-88, 490
Muwatta', 321
mysticism. See Sufism

nabi (prophet), 165

Najran, 194-95

Nakhlah, 175, 176

names, Islamicate, transliteration of.

See transliteration
Nasr II, Samanid ruler, 493
Nasr b. Sayyar, 273
nass, 260, 265, 373
navy

in High Caliphate, 234

in primitive caliphate, 204, 212, 219
Nazwa., in 'Uman, 314
neo-Platonists, 426-28, 439
Nestorians, 140, 157, 306
New Testament, 138, 166, 448
Nihavand, 204
Nile to Oxus region, definition and

usage of term, 60-61, 115
Nishapur, 295-96
Nizaris, 383
nomads. See Bedouin
Nusayris, 383

Nushirvan, Khusraw, 44, 143-44, 201,
282, 304, 448

Occident, definition and usage of, 53-54
Oikoumene, definition and usage of, 50,

109-10
Ottoman empire, 39
Ottoman Turkish, transliteration of. See

transliteration
Oxus, rig, 120, 206, 219, 226, 302

Pahlavi (Parsik) language and literature,
140, 146, 246, 296, 298, 410, 449-50,
4 6 5
and New Persian, 450, 493

Palestine, 115, 198
'pan-Islamic' sentiment, 86
Parthian empire, Parthians, 119, 137,

138, 140, 142, 143
periodization: in Islamic history, 233,

239-40
Persian Gulf, 142, 208
Persian, transliteration of. See

transliteration: Islamicate languages
Persian language and literature. See

Pahlavi language and literature
philosophers. See Faylasuf
Philosophia, 418, 428, 433. See also

Falsafah; Faylasdf
Philosophic tradition, 412, 428-29, 429 n.

See also Falsafah
philosophy. See Falsafah
Piety- minded, 238, 266, 317. See also

Shari'ah-minded; 'ulamd"
and 'Abbasids, 275-80, 283, 299-300,

315. 349
beginnings of, as semi-cultural, semi-
political body opposing ruling trends
among Arabs, 248-56 (see also
Marwani caliphal state. Piety-minded
opposition to)
factions among, 256-67 passim
schism among, 276-78, 371-72

pilgrimage. See hajj

pir {Sufi master), 407

Plato, 117, 136, 239, 426, 429, 430, 435

Plotinus, 426, 429

poetry. See Arabic literature, poetry;
shi'r

postal service, 302

primitive caliphate, 198-230 passim

Ptolemy, 81, 413, 414

qadar, 264

Qadaris (Qadariyyah), 264, 266

qddt (judge in a Shari'ah court), 226, 255,

275. 339. 34 6
Qadi Nu'man, 492
Qadisiyyah, 201
Qahtan, Arab tribal bloc, 229, 245, 267,

271
Qa'im, al-, Fatimid caliph, 492
qandts (underground water channels), 295
Qarmatians, 490-92
qasldah (ode in classical Arabic poetry),

297. 458-59. 460, 464
Qasim al-Rassl, al-, 372
qawl ('word'), 166
Qayrawan (Kairouan), 310, 492
Qays, Arab tribal bloc, 221, 229, 271,

272, 273
qiblah, direction of Mecca, 81

530

INDEX

qiyds, 330, 333, 335

Qubad (Kavad), Sasanian monarch, 143

Qum, 259

Qur'an

assertion of God's unity in, 367-69

on Creation, 165

on divine guidance and prosperity in
household, 185

and family law, 181-83

moral vision in, 163-67 passim

and Muslim piety, 366-71

recitation of, as act of commitment in
worship, 184, 367

references to text of: on God, 185;
on Last Day, 163; on Muslims, 71

as source for Muhammad's life, 160-61
Qur'an reciters, 161, 199, 209, 213, 215,

254
Quraysh, 170-73, 175-77, 187-94 passim,

209, 212, 350
position of, in pre-Islamic Arabia,

154-57
Qusayy, 154-55. *59
qussds (story-tellers), 305, 391
Qutaybah b. Muslim, 226-27

Rabban al-Tabari, 414

Rabbinical Judaism, 125, 316

Rabl'ah, Arab tribal bloc, 229, 273

Rabi'ah, of Basrah, Sufi saint, 402

Ramadan, fast of, 179, 337-38

Raqqah, 300

rationality, rationalism: and Faylas&fs,

422, 424
ra'y, personal judgment in Shart'ah, 329,

33 1 . 333
Rayy, 398, 4°4

Razi, Abu-Hatim al-, Isma'ili da'i, 431-33
Razi, Muhammad b. Zakariyya al- (Lat.

Rhazes), 415, 419, 430, 431-33
Riddah wars, 198, 211, 241, 251, 265
Roman empire, 137-38, 141, 145, 201,

242
R4h (Spirit), 166

Sabi'ans, 298

sacred law. See Shart'ah

Sa'dah, Yemen, 314

Saffah (Abu-1-' Abbas), al-, 274, 284

ijaffarids, 492, 493

sahdbah. See associates of Muhammad

saj' (rhymed prose), 468

Sajid family, 475

saldt (ritual worship), 162, 179, 209, 292,

337. 401
Samanid dynasty, 450, 454, 492-93
Samarqand, 492
Samarra, 470, 472, 482, 486

Sanhajah, Berber tribal bloc, 492
Sanskritic tradition, 103,120
Saragossa, Spain, 310
Sardinia, 138

Sa.sa.nian empire, 43, 153, 191, 192, 194,
198, 199
and absolutist monarchial ideal, 140-41,

280-83, 284, 499
concept of social justice in, 141-42,

143, 282
and Mazdeism, 125, 141-42, 143
Muslim conquest of, 201-6, 212, 227
and Pahlavi language, 138 n, 140,

449-50
populism in, 142-44
taxation in, 242
Sawad, Iraq

under caliphates, 212, 214, 242, 245,

286, 483-85
under Sasanian empire, 201-3, 206
Sayf-al-dawlah, Hamdanid ruler at

Aleppo, 464, 494, 495
Sayf b. 'Umar, 355, 356
Schacht, Joseph, 64
science. See Falsafah; FaylasAf
sect. See firqah
Seleucia, 301
Seleucid dynasty, 119
Shafi'i (Muhammad b. Idris) al-, legist,

64, 326, 327-32, 335, 387, 473
Shafi'i madhhab of legists, 346, 440, 463
shahddah, declaration of faith in Islam,

73. 81, 337, 365
Shaivism, 125
Shari'ah

under 'Abbasids, 345-50
and caliphal court, 347-49
denned, 74, 82, 334 n
development of, by 'ulamd', 318-35
and individual rights and requirements

of public order, 336-44
and Isma'ilis, 379-80
and Sufis, 394-95, 400-402
Shari'ah-minded

and 'Abbasids, 345-50
defined, 238, 318, 351, 351 n
and development of Shari'ah, 318-35
and interest in historical inquiry,
350-58 passim
Sht'is, Shi'ism, 372-74. See also 'Alids;
'Alid-loyalism; imam, imamate;
Isma'ilis; Twelver Shi'is
and 'Abbasids, 272, 274-79 passim,
284, 291, 313-14. 37°. 390-91. 475.
479, 490-92, 493-95
sectarianism among, 260-67 passim
shi'r (poetry), 162, 457-64. See also
Arabic literature, poetry

INDEX

531

shirk, 389

Shurat. See Kharijis

Shu'ubiyyah movement, 461

Sibawayhi of Basra, 297

Sicily, 138, 312, 435, 492

Siffin, 214

Sind, 226, 247

strah, 254

Sistan, 487, 492

Social and Economic History of the Orient,

4 1
Socrates, 429
Soloman, 455

Southern Seas, 142, 151, 233, 314
Spain, 138, 206, 226, 247, 289, 309-10,

335
speculative theology. See kaldm
Stiehl, Ruth, 44

Sufis, Sufism, 238, 266, 371, 393-409
Sulayman, Marwani caliph, 247, 267
Sumeria, 107
Sumerians, no
Sung dynasty, China, 237
sunnah ('established practice'), 252, 275,

27 8 - 324, 325
al-Shafi'l on, 327-32 passim
Sunnls, Sunnism: defined, 276-79, 278 n.

See also Jama'i-Sunnt
siirah, chapter of Qur'an, 162
Susa, 206

Syr-Oxus basin, 140, 143
Syr river, 433
Syria, Syrians

Arab conquest and settlement in,

200-201, 203-4
as centre of Umayyad power, 214,

217-19, 221-23
under Marwanis, 229, 245, 248-51, 253,

256-58, 264, 269, 272-75
Syriac, 140, 146, 235, 239, 410

Ta'abbata Sharran, 460

Tabarl (Ibn-Jarir) al-, 335, 352-57. 391,

455
Tabaristan, 268, 352, 487

Tahart (Tiaret), 310, 314, 492

Tahir, 'Abbasid general, 300, 475

Tahirids, 475-78, 482, 487

Ta'if, 154, 155, 171, 194

Taj Mahall, 63

Talibids, 260

Talmud, 134, 140

Tamim, Arab tribe, 229

T'ang dynasty of China, 233, 237

taqiyyah, pious dissimulation of one's

true opinions, 381
tawhid, assertion of God's unity, 367

Technical Age: definition and usage of,

52-53
Technicalistic. See Technical Age
terminology, scholarly

conventional and revisionist usage of,

45-48
usage of, in Islamicate studies, 55-67
usage of, in world historical studies,

48-55
Thaqif, Arab tribe, 155 212, 223
Thousand and One Nights, 292, 466
Tigris, 138, 202, 274
Toledo, Spain, 3 to
Torah, 448
Toynbee, Arnold, 32
transliteration: Islamicate languages,

7-16
Tulunids, 495
Turks: in 'Abbasid service, 482-83,

485-87
Twelver Shifts, 335, 375~7 8 > 379, 387

Uhud, 190, 192

'ulamd' (the 'learned'; men learned in
Islamic legal and religious studies)
and 'Abbasi regime, 278, 280, 283-84,

289-91. 299-300, 345-50, 473-74, 478
and adibs, 473-74
and kaldm, 443

and ShaH'ah, 238, 280 (see also Piety-
minded; Shari'ah-minded)
'Uman, 194, 314

'Umar b. al-Khattab, caliph, 169, 198,

200, 207-18 passim, 219, 223,

241-42, 243, 245, 251, 270-71

'Umar II (b. 'Abd-al-'Aziz), Marwani

caliph, 258, 268-73, 289, 305
Umayyad amirate in Spain, 309-10
Umayyad caliphate, 212, 217-19, 221 n,

222
Vmmah (community of the faithful).
See also Jamd'ah
and 'Abbasids, 345-50
consolidation of, under 'Umar, 206-1 1
establishment of, in Medina, 172-93

passim
and 'Uthman's death, 354 (see also
Fitnah wars, First).
Upanishads, 450
Urartians, no

Urdu, transliteration of. See transliteration
'ushr (tithe), 270
ttsill al-fiqh, 333

'Uthman (b. 'Affan), caliph, 169, 212, 213,
214, 217-18, 247
al-Tabari on death of, 353-57
'Uthman b. Maz'un, 169

532

INDEX

Utopie und Wirtsckaft: sine geschichtlichte

Betrachtung, 44
'Uzza, al-, Meccan goddess, 156

Vaishnavism, 125
Vedas, 450
vizier, 286, 292
Volga valley, 302

Waardenburg, Jean-Jacques, 28
Waltd, al-, Marwani caliph, 223-26, 247,

249, 267
Waltd II, al-, Marwani caliph, 272
Waqidi, al-, 355-56
Wasit, al-, 223, 245, 386
Wathiq, al-, 'Abbasid caliph, 483, 485
Weber, Max, 31, 35, 105, 133
West. See Occident

Xerxes, 282

Yahya, Barmakid family, 295
Ya'qub b. al-Saflar, 487
Yarmuk river, battle of, 200
Yathrib. See Medina
Yazdagird III of Persia, 202-3

Yazid, 219-21

Yazid II, 271, 272

Yazid III, 272

Yemen, 151-53, 157, 194, 204, 314, 372

Zaehner, Robert, 43-44

zdhir ('externals'), 266

Zahiris, 65

zak&t, 181, 193, 197, 198, 287, 337

Zamzam, well of, 156, 267

Zanj rebellion, 487-88

Zarafshan valley, 226

Zarathushtra (Zoroaster), 115, 128, 141

Zayd b. 'All, 272

Zayd b. Harithah, 167

Zaydis (Zaydiyyah), 314, 372, 473, 487,

494
Zervanism, 44
zindiq, 291
Ziyarids, 494

Zoroaster. See Zarathushtra
Zoroastrianism, 44, 125, 157. See also

Mazdeans, Mazdeism
Zubaydah, wife of Harun al-Rashid, 294,

300
zuhd, 394
Zurvan, a Zoroastrian Dilemma, 43-44